Skip to main content

Full text of "Environmental biochemistry of arsenic"

See other formats


SJVDP L 




ENVIRONMENTAL BIOCHEMISTRY OF ARSENIC 



Prepared Under Contract for the 
Federal -State San Joaquin Valley Drainage Program 



March 1989 



This report presents the results of a study conducted for 
the Federal -State Interagency San Joaquin Valley Drainage 
Program. The purpose of the report is to provide the Drainage 
Program agencies with information for consideration in 
developing alternatives for agricultural drainage water 
management. Publication of any findings or recommendations in 
this report should not be construed as representing the 
concurrence of the Program agencies. Also, mention of trade 
names or commercial products does not constitute agency 
endorsement or recommendation. 



The San Joaquin Valley Drainage Program was established in 
mid-1984 as a cooperative effort of the U.S. Bureau of Reclamation, 
U.S. Fish and Wildlife Service, U.S. Geological Survey, California 
Department of Fish and Game, and California Department of Water 
Resources. The purposes of the Program are to investigate the 
problems associated with the drainage of irrigated agricultural lands 
in the San Joaquin Valley and to formulate, evaluate, and recommend 
alternatives for the immediate and long-term management of those 
problems. Consistent with these purposes, Program objectives address 
the following key areas: (1) Public health, (2) surface- and ground- 
water resources, (3) agricultural productivity, and (4) fish and 
wildlife resources. 

Inquiries concerning the San Joaquin Valley Drainage Program may 
be directed to: 

San Joaquin Valley Drainage Program 
2800 Cottage Way, Room W-2143 
Sacramento, California 95825-1898 



ENVIRONMENTAL BIOCHEMISTRY OF ARSENIC 



Prepared for the 
San Joaquin Valley Drainage Program 
2800 Cottage Way, Room W-2143 
Sacramento, CA 95825-1898 



Under 
U.S. Bureau of Reclamation 
Contract No. 8-PG-20-11780 



By 
S. Tamaki and W. T. Frankenberger, Jr. 
Department of Soil and Environmental Sciences 
University of California, Riverside 
Riverside, CA 92521 



March 1989 



ACKNOWLEDGEMENT 

This literature review was supported by the San Joaquin Valley 
Drainage Program Agreement No. 8-PG-20-11780. We thank Ed Lee, George 
Nishimura and Henry Hansen for their support. 



ENVIRONMENTAL CHEMISTRY OF ARSENIC 

Page 

EXECUTIVE SUMMARY iv 

INTRODUCTION 1 

CHEMISTRY 2 

DISTRIBUTION OF ARSENIC 3 

DETECTION OF ARSENIC IN ENVIRONMENTAL SAMPLES 10 

TERRESTRIAL PLANTS 11 

SOIL MICROBIAL TRANSFORMATIONS 13 

Bacterial Resistance 13 

Bacterial Oxidation 16 

Bacterial Methylation 16 

Fungal Methylation 22 

AQUATIC TRANSFORMATIONS 26 

Marine Environments 26 

Marine algae 26 

Marine invertebrates and fish 31 

FRESHWATER ENVIRONMENTS 36 

MAMMALIAN METABOLISM 38 

TOXICITY OF ARSENIC 42 

STABILITY OF ORGANOARSENIC COMPOUNDS 44 

GLOBAL CYCLING OF ARSENIC 45 

REFERENCES 48 



LIST OF TABLES 

Page 
Table 1. Agricultural use of arsenic 5 

Table 2. Microbial production of alkylated arsines 17 



ii 



LIST OF FIGURES 

Page 

Fig. 1. Oxidation-reduction stability of arsenic 8 

Fig. 2. Anaerobic biomethylation pathway for 

d i methyl a rsine production by Methanobacterium 

sp 21 

Fig. 3. Fungal methylation pathway for the formation 

of trimethylarsine 23 

Fig. 4. Pathways for biosynthesis of arsenobetaine in 

marine algae 30 

Fig. 5. Global biogeochemical cycle of arsenic . 46 



ni 



EXECUTIVE SUMMARY 

Arsenic is a naturally occurring element which has many positive 
agricultural applications in contrast to its biological toxicity and 
threat to wildlife. Arsenicals are used in agriculture as feed additives, 
herbicides, insecticides, and cotton defoliants. The arsenic cycle 
involves both biotic and abiotic reactions. Various plants, invertebrates 
and microorganisms play a major role in the transformation and movement 
of arsenicals in soil and water. 

Arsenic, a metalloid, is naturally present in soil, water, air and 
all living matter. Being in Group V A of the periodic classification it 
forms alloys with various metals and covalently bonds with carbon, 
hydrogen, oxygen and sulfur. Arsenic can occur in the +5, +3, 0, and 
-3 states and is subject to 8 electron reductions which is very similar 
to phosphorus. Because of its similarity to phosphorus and its ability 
to bond with sulfur, it is highly toxic. Arsenate (AsO. ) is taken up 
via the phosphate transport system and is incorporated into energy 
transfer phosphorylation reactions. Arsenite (AsO- - ) inactivates many 
enzymes by having a high affinity for thiol groups of proteins. 

Arsenic, primarily in the inorganic form, is present in the earth's 
crust at an average of 2-5 mg kg" . The mean concentration of arsenic in 
shales is 13 mg kg" ; igneous rocks, 1.8 mg kg" ; and sandstones, 1.0 mg 
kg" . Sedimentary rocks range from 0.1 mg kg to as high as 2900 mg 
kg" arsenic. Sulfidic ores contain arsenic in the form of arsenides of 



TV 



nickel, cobalt, copper and iron. The most frequently found ores are: 
arsenopyrite (FeAsS); enargite (Cu.J\sS 4 ) ; orpiment (As 2 S 2 ); and realgar 
(As.S.). Inorganic arsenic compounds such as arsenic trioxide (As-CL), 
arsenite and arsenate are weathered from arsenic-containing rocks. This 
is considered the major natural source of arsenic estimated to release 
45,000 metric tons/year. The arsenic compounds released as a result of 
weathering may be retained in soils or dissolved in water to be trans- 
ported and further redistributed. The soluble arsenic forms may be 
adsorbed to clay particles in both soils and sediments. 

Precipitation from the atmosphere and the application of 
agricultural products are other major sources of arsenic influx to 
soil. Arsenic deposited from the atmosphere is estimated to be 63,600 
metric tons/year including both wet and dry deposition. Influx of 
arsenic by herbicides to the soil is approximately 4,560 metric tons/year 
and by dessicants, 12,000 metric tons/year. It is estimated that over 
100,000 metric tons of arsenic/year is deposited into landfills as slag, 
a by-product of smelting operations, but only a small percentage of this 
total arsenic enters the global cycle. 

Worldwide, arsenic in soil ranges from 0.1 to 40 mg kg" with a 
median concentration of 6 mg kg" . Arsenic in seawater averages 1.7 ug 
L~ with a relatively homogeneous range from 1.5 to 5 ug L~ . In 
contrast, freshwater from lakes and rivers varies widely in arsenic 
concentration and is dependent upon the minerals subject to transport. 
Freshwater arsenic concentrations range from 1 to 10 ug L" with an 
average of 1.7 ug L~ . The recommended maximum concentration for arsenic 



in irrigation water is 100 yg L~ with the drinking water standard being 
at 50 yg L~ . Arsenate is more predominant in oxygenated water while 
arsenite is more common under reduced anaerobic conditions. The total 
arsenic influx into oceans is estimated at 246,110 metric tons/year. 
Of this total 62,900 metric tons is dissolved arsenic, 178,900 metric 
tons is sediment suspended arsenic and 4,310 metric tons is from the 
atmosphere per year. 

Atmospheric arsenic concentrations are considerably higher over 

land masses in comparison to oceans. The atmosphere over land in the 

-3 -3 

northern hemisphere is approximately 2.8 x 10 yg arsenic m , while 

-3 -3 
over the southern hemisphere, it is estimated at 1 x 10 yg m . 

Atmospheric concentrations of arsenic over oceans is dependent on the 

-4 -3 
proximity to land being 6 x 10 yg m over the North Atlantic and 

-5 -3 
1.8 x 10 yg m over oceans in the southern hemisphere. In suburban 

_3 
areas, air samples have shown arsenic concentrations of 1.7 x 10 yg 

_3 
m with 50% being associated with particles greater than 0.3 microns. 

The alkylarsenic forms comprise an average of 207. of the total atmos- 
pheric arsenic. 

The total arsenic input into the atmosphere has been estimated to 
be 73,540 metric tons/year. Forty percent is from natural sources such 
as volcanism and low temperature volatilization. Sixty percent is from 
anthropogenic sources such as copper smelting, coal combustion, non- 
ferrous metal production, agricultural chemicals and agricultural 
burning. 



VI 



The two commonly used analytical techniques for total arsenic 
determination are atomic absorption spectrometry (AAS) and inductively 
coupled argon plasma emission spectrometry (ICAP) in conjunction with 
hydride generation. Although these two methods are sensitive, they are 
not selective in determining different arsenic species. Analyses by AAS 
or ICAP does not allow direct quantification of arsenate in environmental 
samples but a method was recently developed for the direct determination 
of arsenate in aqueous soil extracts by single-column ion chromatography 
(SCIC) at trace levels. 

In terrestrial plants, arsenate is preferentially taken up 3-4 times 
the rate of arsenite. In the presence of phosphate, arsenate uptake is 
inhibited while in the presence of arsenate, phosphate uptake is only 
slightly inhibited. There is a competitive interaction between arsenic 
and phosphate for the same uptake system in terrestrial plants. The mode 
of toxicity of arsenate is to partially block protein synthesis and 
interfere with protein phosphorylation but the presence of phosphate 
prevents this mode of action. There appears to be a higher affinity for 
phosphate than arsenic with a discriminate ratio of 4 to 1. 

Bacterial resistance to arsenate is related to its chemical simi- 
larities to phosphate and occurs by two distinct mechanisms. Arsenate 
and phosphate are transported into and out of bacterial cells by highly 
specific energy-dependent membrane pumps. One method of arsenate 
resistance in bacteria occurs by activation of a phosphate uptake pump 
with a higher selectivity for phosphate which confers reduced levels of 
arsenate uptake. The second method of resistance is a result of an 



vn 



accelerated efflux of arsenate, arsenite, and antimonate but not phosphate 
from the cell by a highly specific membrane-associated pump. This method 
of resistance for arsenate/arsenite is highly specific in preventing the 
export of phosphate from the cell, and is widespread among different 
bacterial species. Bacterial arsenic resistance is often associated with 
other types of heavy metal and antibiotic resistances. There is a 10-fold 
increase in resistance to arsenate compared to arsenite in bacteria. 

Bacterial oxidation of arsenic from arsenite to arsenate is proposed 
to be a detoxification mechanism. This transformation is induced by 
arsenite, consumes oxygen and is not an energy-yielding reaction. 
Heterotrophic bacteria play an important role in oxidation of arsenite 
to arsenate. 

Methylation of arsenic involves the conversion of inorganic and 
organic arsenic to volatile organic methylated forms such as dimethyl- 
arsine and trimethylarsine. Inorganic arsenic methylation is coupled to 
the methane biosynthetic pathway in methanogenic bacteria and may be a 
mechanism for arsenic detoxification. The pathway proceeds by reduction 
of arsenate to arsenite followed by methylation in the presence of co- 
enzyme M (CoM) , a low molecular weight cofactor found in all methanogenic 
bacteria. Anaerobic biomethylation of arsenic by bacteria proceeds only 
to dimethylarsine, which is stable in the absence of oxygen. In anaerobic 
environments, dimethylarsine can react with disulfide bonds on particu- 
lates in soil thus reducing the concentration of soluble arsenic. In 
general, bacteria are more resistant to methylated arsenic compounds than 
inorganic arsenic species. 



vm 



It has been well known as far back as the 1800s that fungi are 
able to transform inorganic and organic arsenic compounds into volatile 
methylarsines. In Germany and England there were incidences of arsenic 
poisoning caused by molds growing on wallpaper laced with arsenic 
containing pigments producing trimethylarsine, a toxic, garlic smelling 
gas. Since that time, several fungi have been identified as arsenic 
methylators. In the presence of phosphate, biomethylation of inorganic 
arsenic and methanearsonic acid (MAA) is inhibited. 

Arsenate is the primary arsenic species in seawater at 1-2 ug L~ . 
Marine algae partially detoxify arsenate by producing large quantities of 
stable non-volatile methylated arsenic compounds. This is considered to 
be a beneficial step not only to the primary producers, but also to the 
higher trophic levels, since methylated arsenic is much less toxic to 
marine invertebrates. 

Four classes of marine phytoplankton have the ability to absorb 
arsenate and convert it into a reduced and methylated species. These 
are diatoms, coccol ithophorides, dinoflagel lates and green algae. The 
organoarsenic products most commonly excreted from algae are MAA and 
dimethylarsinic acid (DMA). These products are not very stable in natural 
waters since their production and release is balanced by bacterial 
removal by demethylation and oxidation of arsenite to arsenate. 

Marine invertebrates and fish are part of the higher trophic levels 
in the food chain of aquatic environments and often retain 997. of the 
arsenic in the organic form upon consumption. Crustacean and mollusk 
tissues are generally higher in arsenic concentrations than are fish. 



IX 



Arsenosugar and arsenolipids are transformed into arsenobetaine which is 
generally the end product that accumulates in the higher trophic levels. 
Degradation of arsenobetaine is required to complete the biological 
cycling of arsenic. Microorganisms in sediment samples degrade arseno- 
betaine into trimethylarsine oxide, then into DMA and finally to MAA and 
inorganic arsenic. Apparently these derivatives may be subject to vola- 
tilization since the arsenic concentration in these sediments often 
decreases. 

Variability of arsenic in freshwater, ranging from 1 to 10 ug L 
with some estimations as high as 64 ug L~ , depends upon evaporation and 
condensation rates, direct contamination by herbicides or indirectly as 
runoff, industrial pollution and natural contamination. Arsenic concen- 
trations as high as 70 ug L have been reported in lakes of New Zealand 
as a result of hot springs rich in arsenic arising from geothermal 
activity. Hot springs in Yellowstone are as high as 3500 ug of arsenic 
L" . In the Central Valley of California, extensive irrigation has been 
a major factor in mobilizing and redistributing arsenic. Concentrations 
as high as 2400 ug L~ have been reported in evaporation ponds in the 
Tulare Lake Basin of central California. 

At least four species of freshwater green algae including 
Ankistrodesmus sp., Chlorel la sp., Selenastrum sp., and Scenedesmus sp., 
methylate arsenite to MAA and DMA and all, except Scenedesmus , produce 
trimethylarsine oxide. Freshwater algae like marine algae synthesize 
1 ipid-soluble arsenic compounds and do not produce volatile methylarsines. 
Aquatic plants also synthesize similar 1 ipid-soluble arsenic compounds. 



In humans and animals, arsenic enters the body by ingestion and 
inhalation and is removed by first being rapidly absorbed, then assimi- 
lated into the blood followed by removal in the kidneys and excreted in 
the urine. Another possible route is where inorganic arsenic is con- 
verted to methylated forms, which are less toxic and are rapidly excreted 
from the body. 

Arsenic toxicity, characterized by decreased motor skills, nervous 
disorders, respiratory distress and damage to the kidneys, depends on 
its oxidation state. Arsenate breaks down energy metabolism by inhibit- 
ing ATP synthesis by uncoupling oxidative phosphorylation. Arsenite 
inactivates enzymes by linking with sulfur and reacting with thiol groups 
on the active site of many enzymes and proteins. For this reason 
arsenite is not excreted through urine as easily as arsenate making 
arsenite more toxic. Methylation of inorganic arsenic in animals and 
humans produces much less toxic organoarsenic compounds. The LD 5Q for 
DMA in rats ranges from 700 to 2600 mg kg and for MAA 700 to 1800 mg 
kg" compared to inorganic arsenic forms such as potassium arsenite at 
14 mg kg" and calcium arsenate at 20 mg kg" . The arsenic analogues of 
choline and betaine do not bind to thiol groups and are therefore con- 
sidered nontoxic. 

The volatile arsine gases are very toxic to mammals because they 



-1. 



destroy red blood cells (LD™ in rats; 3 mg kg" ). Further studies o 



n 



dimethylarsine and trimethylarsine toxicity by inhalation to test animals 
are still needed. 



XI 



Many organisms including microorganisms, plants and invertebrates 
are involved in the distribution and cycling of arsenic. Arsenic can 
accumulate and be subject to various transformations including reduction, 
oxidation and methylation. The reduced form (arsenite) is considered 
more toxic than the oxidized species (arsenate) because it reacts with 
sulfhydryl groups of cysteine in proteins inactivating many enzymes. 

In aquatic systems, arsenic tends to accumulate as complex organo- 
arsenic compounds with only a few being identified (e.g., arsenobetaine, 
arensochol ine and dimethylarsenosoribosides). MAA and DMA are present in 
seawater and freshwater but appear to be degradation products of these 
complex organoarsenic compounds. 

Arsenic is emitted into the atmosphere by high temperature processes 
such as coal-fired power generation plants, burning vegetation and 
volcanism. Inputs into the atmosphere include industrial and fossil 

8 1 8 1 

fuel emission (780 x 10 g As yr~ ) , mining (28 x 10 g As yr" ) and 

8 1 
continental and volcanic dust fluxes (28 x 10 g As yr" ). 

Natural low temperature biomethylation also releases arsenic into 

the atmosphere. Microorganisms including bacteria, fungi and yeast form 

volatile methylated derivatives of arsenic under both aerobic and 

anaerobic conditions. Bacteria only produce dimethylarsine while fungi 

synthesize trimethylarsine. Dimethylarsine is an oxidation product of 

trimethylarsine and both compounds are subject to demethylation by soil 

bacteria. It is estimated that as much as 210 x 10 g of arsenic is lost 

to the atmosphere in the vapor state annually from the land surface. The 

continental vapor flux is about eight times that of the continental dust 



xn 



flux indicating that the biogenic contribution may play a significant 
role in cylcling of arsenic. It has not been established whether vola- 
tile arsenic can be released by plants. Further studies are needed to 
determine mass balances in the rate of transfer (fluxes) of arsenic in 
the environment. 



XT 1 1 



ENVIRONMENTAL BIOCHEMISTRY OF ARSENIC 



Stanley Tamaki and W. T. Frankenberger, Jr. 

Department of Soil and Environmental Sciences 

University of California 

Riverside, CA 92521 



INTRODUCTION 

The San Joaquin Valley, California is one of the most productive 
agricultural regions in the world because of its climatic conditions. 
However, soils on the west side of the valley contain high amounts of 
soluble salts and trace elements being derived from marine sedimentary 
parent material of the Coastal Range. These fine textured soils contain 
clay lenses which impede water flow and cause shallow water tables. In 
areas of poor drainage, subsurface collectors remove shallow saline 
drainage water to evaporation ponds. Adverse effects have been reported 
on fish and wildlife as a result of selenium contamination at a regional 
evaporation pond facility, Kesterson Reservoir (Merced County), Calif. 

The Coastal Range is the primary source of many potentially toxic 
natural trace elements in the valley drainage water. The elements of 
concern include selenium, arsenic, boron, chromium, mercury, and 
molybdenum. Recently there has been some concern with arsenic occurring 
in the Tulare Lake Basin constituting the southern half of the San 
Joaquin Valley. The Tulare Basin occupies approximately one-third of the 
Central Valley. Elevated concentrations of arsenic in sediments and 
water in the Tulare Lake Drainage District's South Basin evaporation 
ponds may pose a threat to the wildlife. Levels of arsenic in these 



drainage evaporation ponds have been reported as high as 770, 830 and 
2,400 ug L~ at Pyrse, Lost Hills and the Carmel Ranch, respectively. 
The ubiquity of arsenic in the environment, its biological toxicity and 
its redistribution are factors invoking public concern. This report will 
review the chemistry of arsenic and methods currently employed to speciate 
the prevalent chemical forms in various environments. The major focus of 
this paper is on biological transformations of arsenic in both terres- 
trial and aquatic environments. Emphasis will be placed on the link of 
these biotic transformations to the global cycle of arsenic. 

CHEMISTRY 

The arsenic cycle involves both biotic and abiotic reactions. 
Arsenic is a naturally occurring element, present in soil, water, air 
and all living matter. It is a metalloid of Group V A of the periodic 
classification with properties which allow it to form alloys with 
various metals and covalent bonds with carbon, hydrogen, oxygen and 
sulfur (Ferguson and Gavis, 1972). The oxidation states and electron 
orbitals are similar between arsenic and phosphorus. Arsenic is 
subject to eight electron reductions and can occur in +5, +3, and -3 
states. The metal, As, is very rare and is only found under extreme 
redox potentials. Natural sediments and soil produce both nonvolatile 
and volatile methylated arsenic compounds. Its similarity to phosphorus 
and its ability to form covalent bonds with sulfur are two reasons for 
arsenic toxicity. Arsenate (H^AsO.) is an analogue of the essential 
mineral phosphate and is taken up via the phosphate transport system by 



most organisms. Arsenate has been postulated to replace phosphate in 
energy transfer phosphorylation reactions. Arsenite (As0 2 ~) has a 
high affinity for thiol groups of proteins inactivating many enzymes. 

DISTRIBUTION OF ARSENIC 

Arsenic is present in the earth's crust at an average of 2-5 mg kg" 
and is primarily associated with igneous and sedimentary rocks in the 
form of inorganic arsenic compounds. The mean concentration, of arsenic 
in shales, igneous rocks and sandstones is 13, 1.8 and 1.0 mg kg" , 
respectively (Onishi and Sandell, 1955; Lemmo et al., 1983). Sedimentary 
rocks including coal have been found to contain 0.1 to 2900 mg kg" of 
arsenic (Irgolic et al., 1983). It is frequently a component of sulfidic 
ores in the form of arsenides of nickel, cobalt, copper and iron (Irgolic 
et al., 1983). The most commonly found ores are: arsenopyrite (FeAsS) 
(most common and widespread); enargite (Cu^AsS.); orpiment (As-S,); and 
realgar (As.S.). Weathering of arsenic-containing rocks liberates arsenic 
in the form of inorganic compounds including arsenic trioxide, arsenite 
and arsenate. Microorganisms have been shown to increase the rate of 
arsenic release from sulfidic ores by catalyzing the oxidation of sulfide 
to sulfate and ferrous to ferric iron. Weathering of rock is considered 
the major natural source of arsenic, estimated to release 45,000 metric 
ton of arsenic/year (Ferguson and Gavis, 1972). Two additional major 
sources of arsenic influx to soil are precipitation from the atmosphere 
and the application of agricultural products. Wet and dry deposition of 
arsenic from the atmosphere is estimated to be 63,600 metric tons/year. 



Table 1 shows the uses of arsenicals in agriculture including feed 
additives and pesticides. Many of these arsenicals do not persist in 
soils and are thought to be rapidly removed by volatilization and/or 
perhaps leaching (Morrison, 1969; Woolson, 1974). The application of 
herbicides and desiccants to soils is estimated to be 4,560 and 12,000 
metric tons/year, respectively. Deposit of arsenic into landfills as 
slag resulting from smelting operations has been estimated to be over 
100,000 metric tons/year but only a small percentage of the total arsenic 
enters the global cycle. 

Worldwide, the median soil concentration is 6 mg kg" with a typical 
range of 0.1 to 40 mg kg" (Bowen, 1979). The weathered arsenic com- 
pounds may be retained in soils or dissolved in water to be transported 
and further redistributed (Wakao et al., 1988). The soluble forms may be 
adsorbed to clay particles in both soils and sediments. Primary com- 
ponents in soil which adsorb arsenic include porous sesquioxides and 
silico-sesquioxidic complexes. Arsenate sorption increases with 
increasing pH exhibiting a maximum at pH 10.5 (Goldberg and Glaubig, 
1988). Sorption of arsenic to metals may be a more important mechanism 
in immobilization than organic matter (Huang and Liew, 1979). Desorption 
of arsenite is highly dependent on reduction of Fe + to Fe . 

Arsenic concentrations in seawater are relatively homogeneous 
ranging between 1.5 to 5 ug L~ (Sanders, 1980) with an average of 
1.7 ug L~ (Chilvers and Peterson, 1987). In contrast, concentrations 
in freshwater (rivers and lakes) vary widely and are dependent on the 
available minerals subject to co-transport. The average arsenic 
concentration in freshwater is approximately 1.7 ug L (Chilvers 



Table 1. Agricultural uses of arsenic. 



Arsenicals 



Use 



Feed Additives 

Arsanil ic acid and 
Roxarsone (3-nitro-4-hydroxyphenyl 
arsonic acid) 



Carbarsone and Nitarsone 
(4-nitrophenyl arsonic acid) 

Pesticides 

Herbicides (mono-disodium salts of 
methanearsonic acid) 

Insecticides 

Calcium arsenate 



Lead arsenate 

Cotton dessicants 

Dimethylarsinic (cacodylic) acid 
Others 

Sodium arsenite 



Lead and calcium arsenate 
10,10-oxybisphenoxarsine 

Calcium arsenate 



increase rate of gain, and 
improve feed efficiency in 
chickens and swine, control 
swine dysentery 

antihistomonads in turkeys 



post emergence grass herbicides 



control boll weevil in cotton 
fields 

control codling moth, plum 
curculio, cabbage worm, potato 
bug, tobacco hornworm 



cotton defol iants 



Used to preserve railroad and 
telephone posts, fungicide and 
wood preservation. 
Control measles and dead arm of 
table grapes 

control acidity in grapefruit 

fungus control in cotton sail- 
cloth and vinyl films 

component in snail baits; used 
in fly control in poultry 
houses, herbicide for grass 
Poa annua 



and Peterson) with most waters ranging from 1 to 10 yg L~ (Sanders, 
1980). The drinking water standard for arsenic is 50 yg L" and the 
recommended maximum concentration for irrigation water is 100 yg L~ 
(Letey et al., 1986). The arsenate species is most often found in oxy- 
genated water, while arsenite predominates under reduced conditions. 
Particulates of river discharge may contain arsenic concentrations of 
3 to 74 mg kg dry wt (Crecelius et al., 1975). Influx into oceans con- 
sists of 62,900 metric tons of dissolved arsenic, 178,900 metric tons of 
sediment-suspended arsenic, and 4,310 metric tons from the atmosphere per 
year. Thus the total arsenic influx into oceans is estimated at 246,110 
metric tons/year. The major sources of loss from oceans is by sedimenta- 
tion and sea-salt spray leaving a net increase of 136,000 metric tons of 
arsenic/year (Edmonds and Francesconi, 1987). No formation of arsine gas 
has been reported from marine environments. 

Atmospheric concentrations of arsenic are considerably higher over 

land masses with 2.8 x 10" yg m~ reported in the northern hemisphere 

-3 -3 

and 1 x 10 yg m in the southern hemisphere. The atmospheric concen- 
tration of arsenic over oceans is dependent on the proximity to land 

-4 -3 -5 -3 

approaching 6 x 10 yg m over the North Atlantic and 1.8 x 10 yg m 

over oceans in the southern hemisphere (Chilvers and Peterson, 1987). 

Air samples collected in suburban areas have shown concentrations of 

-3 -3 

1.7 x 10 yg m with 50% of the atmospheric arsenic associated with 

particles greater than 0.3 microns. On an average, 20% of the total 
arsenic in the atmosphere is in the alkyl-arsenic form (Johnson and 
Braman, 1975). 

The emission of arsenic into the atmosphere is from natural and 
anthropogenic sources with a ratio of approximately 60 to 40 percent, 



respectively. The total arsenic input into the atmosphere has been 
estimated to be 73,540 metric tons/year. Natural sources include low 
temperature volatilization and volcanism. Copper smelting and coal 
combustion account for 60% of the total anthropogenic source with 
nonferrous metal production, agricultural chemicals and agricultural 
burning accounting for the remaining inventory (Chilvers and Peterson, 
1987). 

To enter the biological cycle arsenic must be in a dissolved form. 
The oxidation and ionization state of arsenic is dependent on the pH 
and oxidation-reduction potential (pE) of the aqueous solution (Lemmo 
et al . , 1983). Figure 1 shows the thermodynamical ly stable forms of 
arsenic in various aqueous conditions. In natural waters, arsenate is 
the predominant species and is in equilibrium with the reduced form, 
arsenite. Dissolved inorganic arsenic compounds may be assimilated by 
organisms and transformed into less toxic methylated derivatives or into 

volatile arsines. Concentrations of dimethylarsinic acid have been 

_i 
reported from 0.05 to 1 ug L in interstitial waters (Crecelius, 1975). 

In oxidized waters, the following arsenic species are stable, 

-2 -3 

HLAsO., H o As0. , HAsO. and AsO. . Arsenous acid species which exists 
3 4 2 4 4 4 

-2 

in reduced waters include H,As0 3 , H-AsO., and HAsO, . Arsine gases are 

only slightly soluble in water and produced mainly in reduced waters 
(Ferguson and Gavis, 1972). Under most conditions, microbially produced 
methylated arsines are rapidly oxidized. Only a small percentage of 



8 9 10 11 12 13 14 



PE 



+ 9 

+8 

+ 7 

♦6 

+5 

+4 

♦3 

+ 2 

♦1 



"1 

-2 

-3 

"4 

-5 

-6 

-7 

'8 

"9 

-10 
"11 

-12 

"13 

-14 



Reduced 
Water 



_AsH 3 (g) 




HAs0 4 * 2 




Oxidation-reduction stability Diagram tor arsenic 



Fig. 1. Oxidation-reduction stability of arsenic. 



methylarsines migrate out of their point source (Lemmo et al., 1983). 

The dissolved forms of arsenic can react to form solids such as As o c 

l b 

and As 2 3 (Ferguson and Gavis, 1972). Under conditions where sulfides 
are stable, arsenic can interact with sulfur to form As.S. (realgar) 
and As 2 S 3 (orpiment) which also have low solubilities and form stable 
solids at pH values below 5.5 and pE values of 0.0 V. 

Adsorption of arsenic onto sediments and coprecipitation with 
ions of sulfur, iron and aluminum are the major mechanisms for arsenic 
removal from aqueous environments. Both arsenate and arsenite co- 
precipitate or adsorb onto hydrous iron oxides. Iron oxides are major 
components of clays and thus removal of arsenic from aqueous solutions 
is dependent on the clay content of the underlying soil or sediment. 
Arsenate competes with phosphate in adsorption to clay surfaces. 
Arsenite is depleted from aqueous solutions due to its strong affinity 
for sulfur and adsorbs or coprecipitates with metal sulfides. Both iron 
and phosphate concentrations present in waters are significant factors 
in establishing the level of dissolved arsenic (Lemmo et al., 1983). 
Arsenic is released from sediments if ferric iron or sulfide is converted 
to ferrous iron or sulfate. However, the final concentration of arsenic 
in aqueous solutions cannot be simply explained by its interaction 
with iron, sulfur and phosphate but other factors such as adsorption- 
desorption equilibria and the total arsenic entrapped in sediments 
also plays a major role (Lemmo et al., 1983). 



10 



DETECTION OF ARSENIC IN ENVIRONMENTAL SAMPLES 

The volatile methylarsines can be detected by gas chromatography 
(GC) with flame ionization detection. This is particularly important in 
speciating arsine gases such as methylarsine, dimethylarsine and 
trimethylarsine. Volatile species can be removed from natural waters and 
soil by gas stripping being collected in a cold trap and separated by GC. 
An element-specific detector such as a GC-arc atomic emission detector or 
a microwave-induced atomic emission spectrometer would be ideal in selec- 
tive detection of organoarsenicals. Mass spectrometric techniques such 
as Fast Atom Bombardment Mass Spectrometry (FABMS) and Field Desorption 
MS could be used to identify organoarsenic compounds after chromato- 
graphic separation. 

Many organoarsenicals are not volatile being quite polar and water 
soluble. Reverse phase-high performance liquid chromatography (HPLC) 
would be more applicable for determination of these nonvolatile organo- 
arsenic compounds. High selectivity could be achieved with a specific 
detector such as a graphite furnace atomic absorption spectrometry (AAS) 
with a deuterum lamp. HPLC-AAS systems have been used to detect 
arsenite, arsenate, methanearsonic acid, dimethylarsinic acid, phenyl- 
arsonic acid, arsenobetaine and arsenochol ine (Craig, 1986). 

The commonly used analytical techniques for total arsenic determina- 
tion are atomic absorption spectrometry (AAS) (Irgolic et al., 1983; 
Owens and Gladney, 1976) and inductively coupled argon plasma emission 
spectrometry (ICAP) (Morita et al., 1981) in conjunction with hydride 
generation (Irgolic et al., 1983). These two methods, though sensitive 



11 



are not selective in direct determination of different species of arsenic. 
Quantitative data for total organoarsenicals is usually limited because 
of the indirect determination calculating the differences between inor- 
ganic species and total arsenic. 

Analyses by AAS or ICAP do not allow direct determination of 
arsenate in environmental samples. Recently a method was developed 
for the direct determination of arsenate in aqueous soil extracts by 
single-column ion chromatography (SCIC) at trace levels (Mehra and 
Frankenberger, 1988). Separation is carried out on a low-capacity 
anion-exchange resin column being quantified by conductometric detection. 

Trace amount measurements of arsenate (detection limit, 92 ug L" ) were 

3- 2- 
made in the presence of other oxyanions, NO., , PO. and SO. . 

TERRESTRIAL PLANTS 

Terrestrial plants growing on shores bordering arsenic contaminated 
water show relatively little arsenic content despite the fact that 
sediments may contain levels as high as 200 ug As g~ (Reay, 1972). 
Furthermore resistance to arsenic can be increased by acclimating plants 
to successively higher concentrations of arsenic. Little bluestem, 
Andropogon scoparius Michx. can survive in soil containing up to 41,200 
mg kg of arsenate (Wauchope, 1982). The soil matrix plays an important 
role in availability of arsenic to plants. Retort oil shales have high 
concentrations of arsenic but plants grown on these soils show little 
accumulation ranging from 0.03 to 0.44 mg kg" (Kilkelly and Lindsay, 
1982). Plant toxicity to arsenic is often reached prior to accumulation 
of levels which would be toxic to wildlife ingesting the plants. 



12 



In studies of arsenate uptake by barley, it was reported that uptake 
is temperature dependent increasing with increasing temperatures (Asher 
and Reay, 1979). Arsenate is preferentially taken up 3-4 times the rate 
of arsenite. The presence of phosphate inhibits the uptake of arsenate 
while arsenate only mildly inhibits phosphate uptake. In terrestrial 
plants as in most other systems, arsenate and phosphate are thought to 
compete for the same uptake system but there appears to be a higher 
affinity for phosphate (Asher and Reay, 1979). The discriminate ratio 
between phosphate and arsenate is approximately 4 to 1. 

In corn, peas, melons and tomatoes, absorbed arsenate is rapidly 
reduced to arsenite. Terrestrial plants do not synthesize 1 ipid-soluble 
arsenic compounds. Methylation of the arsenite occurs under phosphate 
deficient conditions and increases substantially when plants are also 
made nitrogen deficient. The methylated compounds accumulate to approxi- 
mately six times greater concentration in the leaves than in the roots. 
Plants which show high levels of arsenic methylation are atypical and are 
considerably deformed due to the nutrient deficient growth conditions 
(Nissen and Benson, 1982). Synthesis of methylated arsenic compounds by 
terrestrial plants is not necessarily a detoxification mechanism. 

Arsenate toxicity can be detected at levels as low as 1 mg kg~ in 
the roots of All ium cepa , causing a significant reduction in root length 
while 3 mg kg" terminates root growth (Pepper et al., 1968). Inhibition 
of growth is reversible by removal of arsenate. The mode of toxicity of 
arsenate is to partially block protein synthesis and/or interfere with 
phosphorylation of proteins. The addition of phosphate abolishes the 
effects of arsenate. 



13 



MICROBIAL TRANSFORMATIONS 
Bacterial Resistance 

Arsenate is a biochemical analogue of phosphate and is transported 
by highly specific, energy-dependent membrane pumps into the cell of bac- 
teria during assimilation of phosphate (Silver and Nakahara, 1983). In 
Escherichia coli , resistance to arsenic can be achieved by two distinct 
mechanisms: a chromosomal or plasmid encoded system. Chromosomal ly- 
encoded resistance occurs by the activation of a phosphate uptake pump 
with an increased selectivity for phosphate. In bacteria, two phosphate 
uptake systems are present, Pit (inorganic Pi_ transport) and Pst (phos- 
phate specific transport). The Pit system i.s constitutive and does not 
discriminate between phosphate and arsenate having a K for phosphate 
equal to the K. for arsenate (25 uM). During periods of phosphate 
starvation or arsenate toxicity the Pst system is activated and despite 
having an identical K. for arsenate, the reduction in cellular arsenic is 
achieved by the higher affinity for phosphate. The K for phosphate is 
0.25 uM, one hundred times greater affinity than the Pit system (Rosenberg 
et al., 1977). Thus the activation of the Pst system confers higher 
levels of arsenate resistance by virtue of reduced uptake of arsenate 
(Silver and Nakahara, 1983). 

Plasmid-determined resistance is a consequence of an accelerated 
efflux of arsenate from the cell. A highly specific membrane-associated 
pump exports arsenate, arsenite and antimonate but not phosphate from the 
cell (Mobley and Rosen, 1982). Plasmid-encoded resistance is distinct 
from the chromosomal ly encoded Pst system and the level of resistance 



14 



obtained from each system is additive. Modular analysis of the plasmid 
encoded resistance has shown three genes primarily responsible for the 
export function. The genes for resistance are clustered on an R-Factor 
Plasmid called R773 (Chen et al., 1986). ArsA and ArsB genes are 
involved in the export of arsenite and antimonate while ArsC is required 
to confer resistance to arsenate. The oxyanion pump is composed of only 
two proteins, a 63 kd hydrophilic ArsA protein and a 45.5 kd ArsB protein 
(Rosen et al., 1988). The ArjsA gene has been sequenced and the deduced 
amino acid sequence shares homology with several adenylate-binding pro- 
teins such as nitrogenase and the 6-subunit of the mitochondrial ATPase. 
The ArsA protein encodes two distinct adenylate-binding consensus 
sequences which have binding affinity for nucleotides and specifically 
catalyzes the hydrolysis of ATP. The binding of ATP by ArsA is independ- 
ent of the presence of oxyanions, however, the rate of ATP hydrolysis is 
dependent on their presence and is stimulated 5-fold by the addition of 
arsenite and 50-fold with the addition of antimonate (Rosen et al., 1988). 
The ArsA protein is mainly cytosolic but a portion is found sedimented 
within the cell membrane and is thought to complex with ArsB. ArsB is 
found in the inner membrane of E, coli and has been postulated to be the 
portion of the pump responsible for the export of anions from the cell. 
The deduced amino acid sequence of ArsB reveals several regions of the 
protein are potentially transmembrane regions. The 16 kd ArsC 
polypeptide modifies the ArsA-ArsB complex allowing the pumping of 
arsenate. ArsC is not required for the efflux of arsenite or antimonate 
(Chen et al., 1986). The plasmid-encoded resistance for arsenate/arsenite 
- highly specific for oxyanions. It fails to protect E, coli cells from 



15 



concentrations of 0.1 mM phenylarsine oxide or 10 mM Na cacodylate 
(Mobley et al., 1983) but is highly selective in preventing the export 
of phosphate from the cell (Mobley and Rosen, 1982). 

Resistance to arsenic is inducible and recently a fourth gene has 
been identified which regulates the arsenic resistance operon. Plasmid 
encoded resistance results from the activation of an anion-translocating 
ATPase with high selectivity for arsenate, arsenite and antimonate (Rosen 
et al., 1988). 

Plasmid-encoded resistance for arsenate/arsenite is widespread 
among different bacterial species (Nakahara et al., 1977; Smith, 1978; 
Dabbs and Sole, 1988). Hybridization studies indicate that a number of 
arsenate resistant strains of Klebsiella pneumoniae and E. co_H possess 
sequences homologous to the plasmid encoded genes. However, genes 
involved in arsenic resistance are not completely conserved among 
different bacterial strains. In Staphylococcus , three genes are also 
involved in conferring resistance to arsenic, however, sequence analysis 
indicates only ArsB, the gene encoding the transmembrane protein, shares 
homology with sequences of R773 (Silver and Misra, 1988). Furthermore, 
strains of E^ con, K^ pneumoniae and K. oxytoca that are resistant to 
arsenic have been isolated which lack sequences homologous with ArsA, 
ArsB or ArsC genes. One significant feature of the epidemiology of 
bacterial arsenic resistance is the fact that it is often associated 
with other types of heavy metal and antibiotic resistances (Smith, 1978). 



16 

Bacterial Oxidation 

Bacteria are approximately 10-fold more resistant to arsenate 
than arsenite. Bacil lus and Pseudomonas species have been isolated which 
can oxidize arsenite to arsenate. A strain of Alcal igenes faecal is 
obtained from raw sewage was capable of oxidizing arsenite (Phillips and 
Taylor, 1976). Energy is not recovered from the reaction but rather, 
oxidation is proposed to be a detoxification mechanism (Osborne and 
Ehrlich, 1976). Alcal igenes sp. isolated from various soils can oxidize 
arsenite. Biochemical studies on resting cells revealed that the oxida- 
tion process is induced by arsenite. The transformation to arsenate 
consumes oxygen. The use of respiratory inhibitors prevented further 
oxidation of arsenite indicating that oxygen served as the terminal 
electron acceptor. In extreme environments such as acid mine waters, 
arsenic concentrations are as high as 2 to 13 mg L and the major 
inorganic species is arsenite; Oxidation of arsenite by heterotrophic 
bacteria play an important role in detoxifying the envrionment catalyzing 
as much as 78 to 96% of the arsenite to arsenate (Wakao et al., 1988). 

Bacterial Methylation 

Table 2 reveals the microbial diversity of organisms which can 
methylate arsenic into various volatile forms. Bacterial methylation of 
inorganic arsenic has been studied extensively in methanogenic bacteria. 
Methanogenic bacteria are a morphologically diverse group consisting of 
coccal , bacillary and spiral forms but are unified by the production of 
methane as their principal metabolic end product. They are present in 



Ul 

a> 

c 



CO 



■a 



<a 






o 



o 

3 
T3 
O 



<TJ 



a> 




o 


■a 


c 


e 


CD 


ca 


t_ 




a> 


CU 


<»- 


TD 


aj 


•r— 


a: 


l_ 




CO 




o 



1/1 

cz 
o 



c 
o 

o 



■a 

ai 

e 



TO 
O 



-Q 


e 


o 


ui 


1_ 


.»— 


u 


c 


•^ 


ca 


^r 


o> 




l_ 




o 


■ 


o 


CM 


L. 




u 


<u 


•^ 


r— 


21 


J3 




rO 








1—1 


C*1 


r». 


CO 


a> 


o> 



*— 


<a 


o 




3 


4-> 




CD 


"O 




c 


(- 


ra 


CD 




en 


CD 


c 


TO 


CD 


•^ 


^~ 


l_ 


r— 


co 


(O 


o 


-C 







^J- 






CO 






en 






i— 1 


00 


CO 


.£ 


*r 


•* 


U 


en 


en 


c 


«-H 


«-H 


CD 

l_ 


* 


* 


U_ 


• 


• 




i — 


»— 


■a 


rO 


r0 


c 
ra 


4-> 


4-> 




a> 


CD 


■"■ 


■a 


T3 


L. 



17 



ca 



CD CD 



ca 



- 


cz 












T3 








ra 


ra 


ui 


o 












CD 








cz 


r— 


4-> 


■^ 












-* 




!_ 




o 


>. 


<J 


4J 












•^ 




O 




Ul 


X3 


03 


■p— 




CM 








CX 








i_ 


O 


i_ 


T3 


ct- 








if) 




CD 




rd 


u 


4.) 


c 


o 




i/> 


>1 






t_ 


J-> 




^— 


ra 


X 


o 




CL 

1— 
< 


4-> 


-Q 




ul 


o 


rO 




>) 


c_) 


CD 


cj 


ai 


u 






.O 


cz 




.£= 








i_> 


ra 


T3 




e 


m 


CL) 


CD 


4-> 


E 


CD 


o 


c 


u 


CD 




3 


O 


ul 


4-> 


CD 


3 


a) 


•r— 


cu 




4J 


4-J 




S_ 


CM 


1_ 


rO 


e 


•r— 


1_ 


.a 


ui 


C4 


X 


ra 


CM 
f-H 


l_) 


in 


ra 


f— 




T3 


l+- 


o 


CD 


CD 


1 — 




<. 


1 — 


>) 


E 


O 


1 


1_ 


<_ 


CO 


1 


3 


co 


"O 




>)-a 


3 


Ul 


*— 


O) 


Q. 


1 


^—» 


E 


i 


ra 


-C 


.c 


o 


•^ 




r— 


ra 




co 


^ 




m 


CD 


4-> 


4-> 


u 


T3 


"O 


<v 


c 


c 


I 


CD 


4-> 


X 


t_ 


■f— 


CD 


cd 


O 


cz 


<-) 


ra 


•^ 


O 


cj 


ui 


o 


J2 


3 


E 


<_> 


Ul 


ra 



■a 
cu 




Ul 




4-1 




T3 




■a 


T3 


•r— 


CT 


CD 


CZ 


o 


C 


4-> 


ra 


ca 


c 




ul 


CJ 


■^ 


Ul 


3 


•f— 


ca 


.a 


o 


c 


4-> CD 


E 


c 


o 


c -a 


3 


CD 


en 


O i- 


1_ 


ui 


l_ 


CJ X 


CJ 


i- 


ra 


o 




ra 


^— 


E H- 


T3 




>) 


3 L. 


ra 


-C 


_cz 


■■— 4-> 


CD 


J-> 


4-> 


T3 


1_ 


■ r— 


CD 


CD </> 


-Q 


3 


E 


E < 



in 

(_ 
ra 

>) 

4-) 
CD 

e 

Q 





l- CD 




CD 


O C 




> 


■a ••- 




•i— 


O c/) 




4-> 


1_ 




ra 


CJ ra 




> 


• ^" 




•1— 


.— c 




1_ 


«- ca 


aj 


CD 


ra 


c 


T3 


CJ)<<- 


•r— 




o 


Ul 


ul 


en 


1_ 


<c 


C CD 


ra 




o > 


^— 


CD 


1_ -r- 


>) 


r— 


4-> 4-> 


-C 


•f— 


</> ca 


4-1 


4-> 


cj 


CD 


<a 


.c •<- 


e 


i^ 


4-> T3 




o 


i- c 


i_ 


> 


3 f- 


4-> 







CD 












> 












• r— 




I 








4-> 




r~ 








ra 




>) 








> 




JZ. 








•f— 


1_ 


4-> 




d) (U IU 


CD 


c_ 


o 


CD 




c c c 


C 


CD 


T3 


E 




^ -r— 'r— 


•r— 


-a 


o 


•r— 




VI VI 1/1 


Ul 






i_ 




U L- i_ 


!_ 


Ul 


CJ 


4-> 




<t3 fl Tj 


ra 


< 


■t— 






^— ^^ ^^ 


^^ 




1 — 


>) 




>)>)>) 


>> 


CD 


1_ 






-C £Z ^Z 


4^ 


^ 


ra 


_Q 




4-> 4-> 4-> 


4-> 


•!■■ 


en 


ra 


aj 


CD CD CD 


CD 


4-> 




X3 


c 


e e e 


e 


ra 


-C= 


o 


«— 




•^ 


r^ 


4-1 


1_ 


C/l 


i_ t_ t_ 


I_ 


o 


■i — 


Q. 


!_ 


4-> 4-> 4-> 


4-> 


> 


3 


— ' 


ra 



CJ 

ra 

o 

ra 



ra 



ra 



3 
Ul 

CD 

Q 



CO 
O 

en 

cz 

ra 



CD 


Ul 




1_ 


• r~ 




40 


Ul 


^— N 


cu 


CO 


E 


O 


•^« 


3 


• r- 


^~ 


• r- 


1_ 


3 


( 


ra 


ca 


r— 


»^ 


u 


• r— 


3 


•^ 


CJ 


CL, 


> 


• r— 


O 


CD 


c 


o 


(_ 


CD 


oo 


X3 


Q. 









E 




3 




C 




CD 


E 


CJ1 


3 


o 


J-> 


Ul 


ra 


>) 


4-> 


(_ 


o 


J= 


c: 


CJ 





Ul 




3 


1_ 


CJ 


CD 


3 


CD 


rO 


r— 


^— 


CZ 


CD 



E 


E 


ui 


Ul 


3 


3 


3 


3 


•r— 


•r— 


< — 


^™ 


r— 


r— 


i — 


r— 


r— 


r— 


• r— 


•r— 


•r- 


■r— 


cn 


CD 


o 


o 


c_ 


t_ 


•»— 


•f— 


CD 


CD 


cz 


cz 


Q- 


CL 


cu 


cu 


Ul 


Ul 


Q. 


ca- 


<C 


< 



ra 
O) 
-Q 
ra 



O) 

<_) 

c 






05 



(1) 
"O 

c 

rO 
X 



T3 

£Z 

rO 
X 

o 
o 





r0 






n 






r^ 






en 






«— t 




^- 




co 


oo 


M 


rv. 


en 


1_ 


en 


>— i 


01 


<— i 


m 


c 


m 


• 


<o 


en 


r— 


X 


c 


<a 


01 T 


(0 




i — r^ 


3 


4-> 


<C CD 




01 


»-H 


T3 




T3 


C 


c 


C - 


r0 


0) 


n3 J3 




^~ 


ro 


3 


^~ 


x r--. 


ro 


3 


O CD 


JZ 


o 


o —< 


O 



18 



tn 

C 

o 



T3 

C 

o 

C_> 







TJ 


ra 






XJ 






•a — 


E 




-^ 


T3 i- 






•i- <_> 


o 




0) 


i— <_> 






O n3 


I_ 




> 


O ra 






rO 


-rz 




•f— 


<o 


c/> 


aj 


o 


<_> 


<u 


4-> 


(J 


O) 


-i-> 


<j f- 




(-J 


rd 


<J ■»- 


(J 


03 


•«- C 


lO 


n3 


> 


•>- c: 


L_ 


c 


C i- 


ra 


C= 


L. 


c •.- 


3 


aj 


O i/) 


2 


OJ 


a) 


o <s> 


O 


1/1 


lO 1_ 




l/> 


m 


i/i i_ 


tO 


L. 


i- ra 


a> 


1_ 


aj 


1- ro 




ro 


ITJ r— 


-t-> 


rrj 


i_ 


(O • — 


U 




aj >, 


r0 




CL 


<u >, 


•' — 


E 


c .c 


l_ 


l_ 




c -C: 


c 


3 


ra 4J 


j-> 


O) 


TJ 


ra -4-> 


O) 


-f — 


-CZ <U 


</l 


CL 


O 


-C QJ 


uo 


a 


-"-> E 


.Q 


Q. 


O 


-•-> E 


t_ 


o 


<u ..- 


3 


O 


3 


13 


1/5 


E T3 


lO 


u 




E -a 



o 

I T3 

C "0 -r- 

O S_ •.- o 

U O <J ro 
ro 

«*— . <_> 

.C .-■ U •.- 

■4-> I— I •»— C 
O —• C M- 

«- — 'O n 

A IA VI L 

< «- ra 

-M ra — 

C OKU >, 

ai e e .e 

•«- i— ro -»-> 

i- c j: ai 

-i-> •»- 4-> e 

3 ro a> •.- 
c 4-> e -a 



■o 
a> 



■o 
o 



to 

i_ 
ra 



>1 



aj 

E 



0) 


<v 


ai 


B 


c 


c 


■*" 


•r— 


•r— 


V) 


to 


to 


l_ 


l_ 


I_ 


r0 


ra 


ra 




p^ 


f— 


»> 


=^ 


>) 


— 


£2 


x: 


4-> 


-w 


■M 


01 


0J 


01 


E 


E 


E 



■a 

aj 

3 

c: 



o 
u 



<u 



ra 



c 
r0 
en 
t_ 
o 
o 

L. 

o 



rrj 

-a 



ra 



ra 

"O 



ra 

u 



ra 



<_a 



CL 



Q. 

tO -r-l 





to 


E 




3 


to 


•f— 


ra 


^ 


c 


^~ 


o 


•*— 


E 


O 


o 


c 


t_ 


<u 


<u 


Q- 


< i 



19 



large numbers in anaerobic ecosystems, such as sewage sludge, freshwater 
sediments and composts where organic matter is decomposing (McBride et al., 
1978). It has been shown that at least one species of Methanobacterium 
is capable of methylating inorganic arsenic to produce volatile dimethyl- 
arsine. Arsenate, arsenite and methanearsonic acid can serve as substrates 
in dimethylarsine formation. Inorganic arsenic methylation is coupled to 
the CH. biosynthetic pathway and may be a widely occurring mechanism for 
arsenic detoxification. 

Methanobacterium strain M.o.H. cell-free extracts, when incubated 

74 
under anaerobic conditions with [ As]Na 2 HAs0 4 , a methyl donor (methyl- 

74 
cobalamin, CH-XoB.-), H-, and ATP, produced a volatile As-dimethylarsine 

(McBride and Wolfe, 1971). The pathway involves the reduction of arse- 
nate to arsenite with subsequent methylation by a low molecular weight 
cofactor Coenzyme M (CoM). CoM has been found in all methane bacteria 
examined and chemically is 2,2'-dithiodiethane sulfonic acid (McBride et 
al., 1978). Methanearsonic acid added to cell-free extracts is not 
reduced to methylarsine but requires an additional methylation step 
before reduction. However, dimethylarsinic acid is reduced to 
dimethylarsine even in the absence of a methyl donor (McBride and Wolfe, 
1971). Whole cells of methanogenic bacteria under anaerobic conditions 
also produce dimethylarsine as a biomethylation end product of arsenic 
but not heat-treated cells indicating that this is a biotic reaction. 
Furthermore, samples collected from a number of different anaerobic eco- 
systems (anaerobic sewage digestor sludge and rumen from cattle) which 
produced methane also transformed arsenate into dimethylarsine. The 



20 



pathway for anaerobic biomethylation of arsenic differs from fungal 
mediation under aerobic conditions and is shown in Fig. 2. 

Under anaerobic conditions biomethylation of arsenic proceeds on,y 
to dimethylarsine, which is stable in the absence of oxygen, but is 
rapidly oxidized under aerobic conditions. Oimethy.arsine in anaerobic 
environments can react with disulfide bonds present on particulates thus 
reducing the concentration of soluble arsenic. 

interestingly, another study Indicated that resting eel, suspensions 
PSe " d °""" laS and Mcanaenes incubated with either arsenite or arsenate 
under anaerobic conditions produced arsine but no other intermediates 
were found (Cheng and Focht, 1979). Aeronomonas sp. and Flavobaoterlum 
sp. isolated from lake water were capable of methylating arsenic to 
dimethylarslnic add (Wong et a,., 1977,. Flavobaoterlum sp. methylated 
d,methy,arsinic acid to trimethylarslne oxide. Hethylation of arsenic 
is pH-dependent with the highest rates often occurring at pH 3.5 to 5 5 
suggesting that arsenic mobilization from the sediments to the overlying 
water phase is enhanced by acidification (Baker et al., 1983). 
Freshwater algae also produce methanearsonic acid and dimethylarslnic 
acid; however, there is some question on whether biomethylation of 
arsenic in freshwater is a widespread cannon process. 

Bacteria are generally more resistant to exposure of methylated 
arsenic compounds than to inorganic species. Achromobacter navo- 
Mcterium, Nooardia, Pseudomonas , Alcaligenes, Aeromonas and Enterobacter 
isolates were a„ capable of growing ,„ media amended with 100 pg m," 1 
methanearsonic add (Shariatpanah, et a,., 1981). Five of the isolates 



21 



^*? H ?p ^ CH 3' B '2 B '2 ^ H 3 

+5 1 2e +3 v * +3| 

HO-As-OH -^-* As-OH — ^ ^-* HO-As-OH 

II II II 



arsenate arsenite methylarsonic acid 



CH 2 -B, 2 B, 2 CH 3 4e CH 3 

> HO— As-CH_-^ — > As-CH, 

/ II ^ | 3 



2e 



H 

dimethylarsinic acid dimethylarsine 



Fig. 2. Anaerobic biomethylation pathway for dimethylarsine 
production by Methanobacterium sp. (McBride and 
Wolfe, 1971). 



22 



Achromobacter , Flavobacterium , Nocardia, Pseudomonas , and Alcal igenes 

14 14 
possessed a demethylating enzyme which released CO- when C-methane- 

arsonic acid was added as a substrate. Demethylating activity has also 

been reported in two isolates of Actinomycetes (Von Endt et al., 1968). 



Fungal Methylation 

It is well established that fungi are able to transform inorganic 
and organic arsenic compounds into volatile methylarsines. The volatil- 
ized arsenic dissipates from the cells effectively reducing the arsenic 
concentration the fungus is exposed to. The importance of fungal meta- 
bolism of arsenic dates back to the early 1800' s where a number of 
poisoning incidents in Germany and England were caused by a volatile 
methylarsine gas. The victims lived in musty rooms with a characteristic 
garlic-like odor. Trimethylarsine was identified as the toxic compound 
(Challenger, 1945). Molds growing on wallpaper decorated with arsenical 
pigments (Scheele's green and Schweinfurter green) produced the toxic 
trimethylarsine gas. Since then, several species of fungi have been 
identified that are able to volatilize arsenic (Cox and Alexander, 
1973a). The fungus, Penicil 1 ium brevicaule ( Scopulariopsis brevicaul is ) 
produces trimethylarsine when grown on bread crumbs containing either 
methanearsonic acid or dimethylarsinic acid. A biochemical pathway for 
trimethylarsine production has been proposed by Challenger (1945) (Fig. 3) 

In recent studies, three different fungal species Candida humicola , 
Gl iocladium roseum and Penici 1 1 ium sp. were capable of converting 



23 



2e" 







2- 



H 2 As v O~ 



arsenate reduction 
2e" O 2 " 



As m O(OH)- 
arsenite 

0" 

1 ill 
CH 3 -As m =0 

methanearsonous anion 



CH + 



H 



OH 

ChL-As v =0 
3 I 
0" 



methylation methanearsonic acid 



CH + 



H' 



^LJ> 



2e" O 2 " 

^-4 



2e' 







2- 



CH 3 
CH 3 -As m -0" 

dimethylarsinous anion 
CH, 



CHt H 



±~4> 



i 



in 



:As M, -CH, 

I 3 

CH 3 

trimethylarsine 



CH 3 

I * 

CHL-As v =0 
5 i 



dimethylarsinic acid 

CH 3 

CH^-As v =0 
3 l — 

CH 3 
trimethylarsine oxide 



Fig. 3. Fungal methylation pathway for the formation of 
trimethylarsine (Craig, 1986 [modified after 
Challenger, 1945]). 



24 



methanearsonic acid and dimethylarsinic acid to trimethylarsine (Cox and 
Alexander, 1973a). In addition, C^ humicola used arsenate and arsenite 
as substrates to produce trimethylarsine. Cell-free homgenates of C^ 
humicola transformed arsenate into arsenite, methanearsonic acid and 
dimethylarsinic acid (Cullen et al., 1979). Although methylation of 
inorganic arsenic and methanearsonic acid is inhibited by the presence of 
phosphate, the rate of trimethylarsine formation from dimethylarsinic 
acid is increased in the presence of phosphate (Cox and Alexander, 1973b). 

Methylation of arsenic is thought to occur via transfer of the 
carbonium ion from S-adenosylmethionine (SAM) to arsenic. Incubation of 
cells with an antagonist of methionine inhibits production of arsines 
thus supporting the role of methionine as a methyl donor (Cullen et 
al., 1977). The addition of either methanearsonic acid or dimethylarsinic 
acid to cell-free-extracts yields trimethylarsine oxide (Cullen et al., 
1979). Further reduction of trimethylarsine oxide to trimethylarsine 
requires the presence of intact cells (Pickett et al., 1981). Various 
arsenic thiols (cysteine, glutathionine and lipioc acid) are thought to 
be involved in the reduction step of trimethylarsine oxide to trimethyl- 
arsine (Cullen et al., 1984a; 1984b). The final reduction step is 
inhibited by several electron transport inhibitors and uncouplers of 
oxidative phosphorylation (Zingaro and Bottino, 1983; Pickett et al., 
1981). Preincubation of cells with trimethylarsine oxide increases the 
rate of conversion to trimethylarsine suggesting an inducible system 
(Pickett et al., 1981). In addition, the rate of transformation of 
arsenate to trimethylarsine is increased by preconditioning the cells 
with dimethylarsinic acid (Zingaro and Bottino, 1983). The compounds 



25 



isolated during the reduction of arsenate by C^ humicola is consistent 
with the intermediates reported in the pathway for methylation of arsenic 
as proposed by Challenger (1945). 

Soil fungal species may play a major role in the transformation and 
movement of arsenic chemicals used in agriculture. The methylation of 
arylarsonic acids is important because of their wide use as food supple- 
ments for swine, turkeys and poultry. Candida humicola is capable of 
methylating benzenearsonic acid to produce volatile dimethylphenylarsine 
(Cullen et al., 1983). In addition, methylphenylarsinic acid and 
dimethylphenylarsine oxide are also reduced by C^ humicola to 
dimethylphenylarsine. Arsanilic acid, which contains an amino group at 
the para position of phenylarsonic acid was not converted to a volatile 
arsine but it has been reported that soils treated with arsanilic acid 
can lose its arsenic component. 

The adaptiveness of C^ humicola in methylating arsenic is evident by 
the fact that dilute solutions of the highly effective wood preserving 
fungicide, chromated copper arsenate (CCA), is depleted of arsenic through 
volatilization (Cullen et al., 1984). It has also been demonstrated that 
a variety of soils have the potential to produce alkylarsines (Woolson, 
1977). Soils amended with inorganic and methylated arsenic herbicides 
produce dimethylarsine and trimethylarsine (Woolson et al., 1973; Baker 
et al., 1983; Hassler et al., 1984; Woolson, 1977). The organisms 
responsible for volatilization are from diverse environments suggesting 
that a number of different species have the capacity to produce 
alkylarsines. Mixed communities of microorganisms in soil produced 
dimethylarsine and trimethylarsine in headspace trapped in bell jars 



26 



over soil and lawn treated with methyl arsenical s (Braman and Foreback, 
1973). 

Arsenic biomethylation is also widespread among higher organisms and 
has been demonstrated in human urine, bird eggshells, cows, dogs, rats, 
mice, rabbits, freshwater fish (trout), and terrestrial plants 
(tomatoes). Higher plants such as pine, corn, melons (honeydew) and pea 
reduce arsenate to arsenite but do not produce organoarsenic compounds 
(Nissen and Benson, 1982). Among the animals, there is some question on 
whether the intestinal bacteria are responsible for arsenic methylation. 

AQUATIC TRANSFORMATIONS 

Marine Environments 

Marine algae . In seawater, arsenate is the predominant arsenic 
species and is present at approximately 1-2 ug L (ppb) (Andreae, 1979). 
In highly productive environments, the phosphate concentration can be 
depleted to levels below the concentration of arsenate (Benson et al., 
1981). Arsenate is assimilated due to its similarity to phosphate by 
marine phytoplankton (Sanders, 1979) as evident with uptake studies in 
both bacteria (Silver and Misra, 1988) and marine yeast (Button, 1973). 
However, phosphate and arsenate uptake may be independent processes in 
marine phytoplankton suggesting a non-competitive absorption mechanism 
(Andreae and Klumpp, 1979; Klumpp, 1980). Regardless of the mechanism 
of arsenate uptake, primary producers must adapt to the accumulation of 
cellular arsenate. Arsenate is partially detoxified by production of 
large quantities of stable methylated arsenic compounds. These 
methylated compounds do not interfere with phosphate esterif ication. 



27 



Approximately 15 to 20% of the total soluble arsenic in marine biota 
systems are reduced and methylated during uptake (Sanders and Windom, 
1980). The seasonal increase and decrease of arsenic in marine waters 
is consistent with the fluctuation in biological activity (Howard et al., 
1982). The formation of methylated arsenic compounds is considered as a 
detoxification step beneficial not only to the primary producers, but 
also to higher trophic levels, since these compounds are much less toxic 
to marine invertebrates (Sanders, 1979). 

The ability to transform arsenate to arsenite and subsequently to 
methylated arsenic compounds is present in four classes of marine 
phytoplankton: diatoms, coccolithophorids, dinoflagellates and green 
algae (Prasinophyceae). Arsenate is absorbed and converted to arsenite, 
methanearsonic acid and dimethylarsinic acid before being secreted from 
the cells (Andreae and Klumpp, 1979). A portion of the arsenate, 
however, is retained within the cell and metabolized further into complex 
organic compounds. The metabolism of arsenic is species dependent but in 
general, diatoms, dinoflagellates and green algae convert 30-50% of the 
retained arsenic into 1 ipid-soluble compounds while coccolithophorids 
incorporate less than 1% (Andreae and Klumpp, 1979). 

The chemical form of arsenic in marine waters influences the cellular 
concentration and metabolism of arsenic. Enrichment of media with high 
levels of arsenate increased the organic arsenic pool in Skeletonema 
costatum (Andreae and Klumpp, 1979). Arsenite invoked increased levels 
of arsenic in both the organic and inorganic cellular fraction. The 
uptake of arsenate was four times the rate of arsenite in Fucus spiral is 
(Klumpp, 1980). 



28 



Enrichment of marine phytoplankton cultures with dimethylarsinic 
acid does not increase the cellular arsenic concentration or change the 
inorganic/organic arsenic ratio (Sanders and Windom, 1980). Furthermore, 
the addition of dimethylarsinic acid does not affect diatom productivity 
(Sanders, 1979). Thus it has been suggested that the conversion of 
arsenate to methanearsonic acid or dimethylarsinic acid may possibly be 
a detoxification mechanism where the metabolism of arsenic into lipid- 
soluble compounds may in fact be an adaptation by these marine organisms 
to compensate for limited nitrate availability. The periodic table 
indicates that nitrogen and arsenic, both being from group V A, have 
similar chemical properties. It is known that arsenic can replace 
nitrogen in choline used to build arsonium phosphatides which functions 
efficiently as structural lipids (Wrench and Addison, 1981). 

The wide variation in total arsenic among macroalgae species is often 
a reflection of the difference in the size of the organic arsenic pool. 
The arsenic concentration in three main classes of macroalgae ranged 
between 0.4 to 32 ug g" dry weight (Sanders, 1979). The distribution of 
arsenic in the organic pools were as follows: Phaeophyceae (brown algae), 
78%; Rhodophyceae (red algae), 57%; and Chlorophyceae (green algae), 53%. 
The percentages of inorganic arsenic varied widely between species, both 
within and among groups but the actual concentrations were within a 
narrow range of 0.63 to 2.46 ug g~ . Arsenate assimilation by microalgae 
is most likely a function of phosphate uptake. Species belonging to the 
Class Phaeophyceae contain high levels of phosphate and are thought to 
have high rates of organoarsenic metabolism to compensate for the high 
levels of internal arsenate. 



29 



The chemical structure of several water soluble organoarsenic com- 
pounds (dimethylarsenosugars) have been identified from the brown kelp, 
Ecklonia radiata (Edmonds and Francesconi, 1981; 1983). Figure 4 illus- 
trates the pathway for the biosynthesis of arsenosugars in marine algae. 
The total cellular arsenic concentration can range as high as 10 ug g" 
fresh weight with water-soluble compounds accounting for approximately 
81% of the total arsenic. In contrast, a related brown algae, Fucus 
spiral is contained a single lipid compound which accounted for 60% of 
the total cellular arsenic. Arsenolipid anabolism proceeds through water- 
soluble organic intermediates from arsenate (Fig. 4). The conversion of 
the water-soluble intermediates to 1 ipid-soluble compounds is dependent 
on respiration rather than photosynthesis. Blockage of respiration leads 
to the accumulation of a water-soluble organoarsenics. Chromatographed 
cell extracts of Skeletonema revealed 12 distinct water-soluble 
organoarsenic compounds (Andreae and Klumpp, 1979) but it has not been 
determined if the metabolic pathway is related to that reported in E^ 
radiata (Klumpp and Peterson, 1981). 

The tolerance of marine algae to high concentrations of arsenic is 
exemplified by Tetraselmis chuii (a green flagellate, Chlorophyta) . T\ 
chuii can be acclimated to survive in arsenic concentrations as high as 
1 g L (1,000 ppm) of arsenate by synthesizing arsenolipids with an 
arsenocholine moiety (Bottino et al., 1978a). Highly resistant strains 
are physiologically altered and transfer of these strains to arsenate 
free-media often results in cell death (Bottino et al., 1978b). The 
synthesis and accumulation of organoarsenic compounds within the cell is 
proportional to the external concentration of arsenic in the surrounding 



fO 

CJ 



30 



X v> 

o-< 



So 



t 



fO 

X 
cj 

o— < = o 



So 



t 



ro 

X 

o 

X ' 

o-< 



O 

X 

t 



fO 

X 

o 

I 

X w 

o— < = o 

I 
o 

X 



t 

X 

o 



o-< 






X 








o 








o 
o 

cm 




a> 




X 




o 




o 






rO 

X 


1 


rO 

X 


1) 
-Q 

o 


o- 


1 


o 


c 

<D 

to 




o 




L. 




fO 




o 




X 










I 



/ \ 




X 
O 

o 

CJ 

CM 

X 

cj 

ro | 
X <o 

cj — <— cj 



o 

a 

JO 

o 
a 

>» 

rO U 

X 5 

X 

O 



CD 

E 



X 

o 

cvj 

X 

o 

CM 

X 

cj 

rO | ro 

X w x 
cj— < — cj 

I 

So 

X 



<u 



D 



c 


ai 




o 


o 


e 


r* 


0) 


-^- 


(/> 


a 


i_ 


o 


« 


c 




CD 

to 


<4- 

o 



\ h 



o 

X 


ro 

X 


J — To 
J2 X 


t 


o- 


-<-o 




II 


1 




O 


X 

o 

S « 




{1 


o— < = o 
1 




to 

-a 


o 




'a. 


X 




'—. 









c 




X 




o 




o 




-C 




CM 

X 




_CD 




a 




>% 




CM 




to 




X 




v_ 




CJ 




o 


ro 


1 


rO 


X 


X 


</> 


X 


o- 


■<- 


o 


>> 




II 




-C 










o 




CD 

£ 



1/1 

ai 



</» 

O 



3 



ITS 

a. 



en 



O 

C 
CD 

to 



31 



environment. High concentrations of arsenic are often found in marine 
algae collected along an increasing arsenic gradient at a sea-estuary 
interface (Klumpp and Peterson, 1979). 

Among the organoarsenic compounds synthesized from inorganic arsenic 
in the aquatic environment, methanearsonic acid and dimethylarsinic acid 
are products commonly excreted, but make up a very small fraction of the 
organoarsenic pool in algae. They appear to be intermediates in the 
synthesis of other organoarsenic compounds, including arsenol ipids. 
Although methylarsenic acids are the major forms of organic arsenic 
excreted by algae, they are not very stable in natural waters. Pro- 
duction and release is apparently balanced by bacterial removal such as 
demethylation and subsequent oxidation of arsenite to arsenate. It has 

been estimated that the production rate and removal rate of dimethyl - 

-3 -1 
arsinate are the same in seawater at about 1 ng dm day (Sanders, 

1979). 

Marine invertebrates and fish . Fish and invertebrates are part of 
the higher trophic levels in the food web of marine environments. The 
primary producing phytoplankton and macroalgae can accumulate arsenic 
and transform inorganic forms into complex organic molecules that may be 
converted into water-soluble or lipid-soluble arsenic compounds. 
Consumption of primary producers by a higher trophic level metabolizes 
the organoarsenic compounds into other forms. Fish and marine inverte- 
brates retain 99% of the arsenic in the organic form. The concentration 
of inorganic arsenic rarely exceed 1 mg kg" in their tissues (Maher, 
1983). Generally, arsenic concentrations are higher in crustacean and 
mollusk tissues compared to fish (LeBlanc and Jackson, 1973; Maher, 1983). 



32 



Chemical structures have been determined among several of the 
predominate water-soluble arsenic compounds in invertebrates and fish. 
Giant clams of the genera Tridacna and Hippopus concentrate arsenic in 
their kidneys to levels greater than 1000 mg kg" dry weight (Benson and 
Summons, 1981). The arsenic is obtained from symbiotic unicellular 
algae, zooxanthel lae, which transforms arsenate in seawater to organic 
compounds that are passed to the host clam. In Tridacna maxima kidney 
tissue, two organoarsenic compounds were isolated and identified as 
trimethylarsoniumlactate and O-glycerophosphoryl-trimethylarsoniumlactate 
accounting for the majority of the cellular arsenic (Benson and Summons, 
1981). The most predominant organoarsenic species, however, was not 
identified but was also thought to be a lactate derivative. The synthe- 
sis of arsenolipids are proposed as an adaptation by the clams to reduce 
the cellular arsenic concentration. The arsenolipids protrude from the 
gill membranes into the surrounding marine waters and become accessible 
to bacteria which oxidize the lipids and release the arsenic moiety as 
dimethylarsinic acid (Benson and Nissen, 1982). Dimethylarsinic acid and 
minor amounts of methanearsonic acid are also present in clam tissue 
(Benson and Summons, 1981). 

The unequivocal identification of the water-soluble lactate 
compounds in clam kidney tissue is subject to dispute (Edmonds and 
Francesconi, 1987). Edmonds and Francesconi (1982) isolated and 
determined the X-ray crystal structure of two water-soluble arsenic 
compounds present in the kidneys of Tridacna maxima . These two compounds 
accounted for 80% of the total arsenic in the kidneys and were identical 
to the dimethylarsenosugars isolated from Ecklonia radiate (Edmonds and 



33 



Francesconi, 1981). Arsenosugar synthesis is not unique to E^ radiata 
and their presence in clams was thought to be due to the arsenosugar 
synthesis by the symbiotic algae with passage to the host accumulating in 
the kidneys. However, current evidence suggests that the arsenosugars 
were misidentified as arsoniumlactate compounds by Benson and Summons 
(1981) and thus any reference in the literature to arsoniumlactate com- 
pounds must be carefully evaluated. 

Arsenobetaine has been isolated and purified from the Australian 
rock lobster, Panul irus longipes (Edmonds et al., 1977; Cannon et al., 
1981), and the dusky shark, Carcharhinus obscurus , (Kurosawa et al., 1980; 
Cannon et al., 1981). In flounder, sole, lemon sole, dab, and shrimp, 
the major arsenic species was recovered as arsenobetaine (Luten et al., 
1983). The concentrations of arsenic ranged from 0.45 to 31.4 ug g" dry 
weight. In shark, arsenobetaine is the predominant arsenic species in 
both muscle and liver tissues (Kurosawa et al., 1980). In the muscle, 
807. of arsenic was detected in the aqueous fraction, 10% in the residue 
and almost none in the lipid fraction. In contrast, in the liver, 40% of 
the total arsenic was found in the aqueous fraction while 45% in the 
lipid fraction. In shrimp, arsenobetaine constitutes two-thirds of the 
arsenic pool while the remainder has tentatively been identified as 
arsenocholine (Norin et al., 1983; Norin and Christakopoulos, 1982). 
However, other investigators have been unable to confirm the presence of 
arsenocholine in other shrimp species (Luten et al., 1983; Shiomi et al., 
1984). 

It is unclear how arsenosugars and arsenolipids are transformed into 
arsenobetaine within the higher trophic levels of the marine environment. 



34 



The American lobster, Homarus americanus , retains arsenic in its muscle 
tissue as arsenobetaine (Edmonds and Francesconi , 1981) but is unable to 
synthesize arsenobetaine from organoarsenic compounds obtained from 
ingested algae (Cooney and Benson, 1980). The generation of arsenobetaine 
from arsenosugars requires cleavage at the C^-C. bond of the sugar residue 
and the subsequent oxidation of the C. carbon. Furthermore, reduction 
and methylation of the arsenic atom must occur to form the quaternary 
(tetraalkylated) arsonium compound of arsenobetaine (Edmonds and 
Francesconi, 1987). Dimethylarsinoylethanol (see Fig. 4 for chemical 
structure), a product of anaerobic bacterial decomposition of arseno- 
sugars in Ecklonia , has been proposed as an intermediate in the formation 
of arsenobetaine (Edmonds et al., 1982). However, the process in which 
dimethylarsinoylethanol is transformed into arsenobetaine is unclear. 
There is no evidence that suggests bacteria have the potential to carry 
out the necessary quaternary methylation (Edmonds and Francesconi, 1987). 
Thus the pathway involved in the synthesis of arsenobetaine remains 
largely undefined. 

The synthesis of organoarsenic compounds in marine environments is 
principally by primary producers. Although all of the intermediates of 
metabolism have not been identified, arsenobetaine is thought to be the 
end-product which accumulates in marine animals (Norin and Christakopoulos, 
1982). There is evidence, however, that suggests some marine animals are 
able to synthesize organoarsenic compounds directly from arsenate. The 
macroalgae feeding snail, Littorina littoralis , converts arsenate in 
seawater, into a single water-soluble compound identified as WS-0.66, 
which is distinct from arsenobetaine and arsenochol ine (Klumpp and 



35 



Peterson, 1981). Interestingly, WS-0.66 is also the major arsenic com- 
pound accumulated by snails which feed on the macroalgae, Fucus spiral is . 
The macroalgae store arsenic predominantly as a single 1 ipid-soluble com- 
pound which is presumed to be metabolized by the snails and stored as 
WS-0.66. Nucella lapillus , a predatory snail, also produces and stores 
arsenic as WS-0.66 (Klumpp and Peterson, 1981). Like Littorina 
littoralis , Nucella lapillus is able to convert arsenate obtained from 
seawater or various organoarsenic compounds derived from its food sources 
into WS-0.66. In both snail species, 1 ipid-soluble compounds can be 
recovered but are usually less than 107. of the total arsenic retained. 

Arsenobetaine is the end product which generally accumulates in the 
higher trophic levels (Norin and Christakopoulos, 1982) and its degrada- 
tion is required to complete the biological cycling of arsenic. Microbial 
degradation of arsenobetaine occurs in coastal water sediments. Sediment 
samples amended with arsenobetaine contain organisms which are capable of 
degrading arsenobetaine into trimethylarsine oxide (Kaise et al., 1985). 
Trimethylarsine oxide is further broken down into dimethylarsinic acid 
and finally to methanearsonic acid and inorganic arsenic. The breakdown 
derivatives are apparently volatilized by microorganisms since the con- 
centration of arsenic in the culture media steadily decreased. However, 
it was demonstrated that arsenobetaine as a sole carbon source could not 
support the growth of pure cultures responsible for degradation and thus 
cometabolism may be involved (Hanaoka et al., 1987). 



36 



FRESHWATER ENVIRONMENTS 

Freshwater bodies are more variable than the marine environment in 
the concentration of arsenic ranging between 1 to 10 yg L" (Sanders, 
1980). Other estimations are as high as 64 yg L~ (Schraufnagel , 1983). 
Arsenic concentrations in freshwater fluctuates depending on evaporation 
and condensation rates. Other factors which contribute to variations 
are contamination by arsenic herbicides directly or as runoff, industrial 
pollution and natural contamination. In lakes of New Zealand of the 
Taupo-Wairakei region, the concentration of arsenic has been reported to 
reach concentrations of 70 yg L~ as a result of arsenic-rich hot springs 
arising from geothermal activity (Reay, 1972). Irrigation of farmland in 
the Central Valley, California is a significant factor in mobilizing and 
redistributing arsenic. In the Tulare Lake Drainage District, Kings 
County, California, arsenic concentrations in evaporation ponds have been 
reported as high as 2400 yg L" . Yellowstone hot springs are as high as 
3,500 yg of arsenic L (Stauffer and Thompson, 1984). 

The resistance to arsenate among freshwater algae is highly 
species dependent. Five taxonomically divergent algae, Chlamydomonas 
reinhardtii , Melosira granulata , Ochromonas val lesiaca , Anabaena 
variabil is and Cryptomonas erosa showed a wide range in susceptibility 
to arsenate (Planas and Healey, 1978). At arsenate concentrations of 
75 yg L~ , the growth rates of M^ granulata and 0^ vallesiaca were 
depressed by approximately 20 to 407.. Inhibition in growth by C^ 
reinhardtii was not evident until the concentration of arsenate was 
750 yg L~ and A^ variabilis and C^ erosa both were unaffected by 



37 



concentrations as high as 7,500 yg L . These differential resistance 
levels may play a significant role in species succession. 

Arsenic is metabolized into various methylated forms by freshwater 
algae. Arsenite is methylated by at least four freshwater species of 
green algae, including Ankistrodesmus sp., Chlorel la sp., Selenastrum 
sp., and Scenedesmus sp. (Baker et al., 1983). All four species methyl- 
ated arsenite when present in media at 5,000 yg L~ , approximately the 
same level of arsenite used to control aquatic plants in lakes (Hood and 
Associ, 1985). The levels of recovered methylated arsenic species was 
quite high on a per gram dry weight basis. Each of these organisms 
transformed arsenite to methanearsonic acid and dimethylarsinic acid and 
all, except Scenedesmus , produced detectable levels of trimethylarsine 
oxide. Unlike fungi, volatile methylarsines were not produced (Baker et 
al., 1983), but instead, limnetic (freshwater) algae like marine algae 
synthesize 1 ipid-sol uble arsenic compounds. Freshwater algae grown in 
media amended with 1 to 3 yg L of arsenate synthesized 1 ipid-soluble 
arsenic compounds to levels approximately equal to marine algae (Lunde, 
1972; Lunde, 1973). 

Aquatic plants also have the biosynthetic machinery to synthesize 
arsenolipids, with as much as 50 to 80% of the metabolized arsenic 
converted into a 1 ipid-soluble form. The chemical structures of these 
compounds have not been determined but share chemical characteristics 
similar to those synthesized by the marine algae (Benson et al., 1981; 



Benson and Nissen, 1982). The aquatic dicot, Ceratophyllum demersum , 
can accumulate arsenic concentrations up to 650 yg g~ dry weight when 
isolated from arsenic enriched waters (Reay, 1972). Other aquatic plants 



38 



isolated from New Zealand's Taupo-Wairakei area have been found to 
accumulate high concentrations of arsenic. Aquatic plants isolated from 
these enriched waters have considerably higher concentrations of arsenic 
than the identical species isolated from noncontaminated lakes (Fish, 
1963; Reay, 1972). 

Arsenate given orally to fresh water brown trout ( Salmo trutta) is 
converted to an organic form mediated by the intestinal microflora and 
then rapidly absorbed (Penrose, 1975). Arsenate injected intramuscularly 
is initally detected in the blood as inorganic arsenic but is slowly 
converted to an organic form. Both inorganic and organic arsenic com- 
pounds accumulate in the liver and are secreted with the bile into the 
lumen. Inorganic arsenic is thought to be methylated by the intestinal 
microflora in the gastrointestinal tract and then preferentially reabsorbed 
into the body of the fish. Two organic arsenic compounds were detected 
in trout and were clearly distinct from arsenobetaine. There is no 
clear evidence that fish are capable of directly converting inorganic 
arsenic into methylated species. 

MAMMALIAN METABOLISM 

In humans and animals, arsenic enters the body mainly by ingestion, 
inhalation and sorption and is rapidly excreted or metabolized into less 
toxic methylated organic compounds. Extensive studies have been made on 
the metabolism of arsenic compounds with rats (Coulson, 1935). Rats fed 
arsenic trioxide accumulated 55-65 times more arsenic than control rats, 
whereas those fed equivalent amounts of shrimp-derived arsenic (now known 
as arsenobetaine), retained only 2-4 times the amount of the controls. 



39 



Approximately 807. of the ingested arsenic trioxide was retained in the 
body compared to. less than 2% of the fed shrimp-arsenic. The shrimp- 
arsenic was almost exclusively excreted in the urine. In contrast to 
rats, monkeys fed fish-arsenic (also known as arsenobetaine) , excreted 
57 to 84% of the ingested arsenic in the urine within 4 days, with little 
found in the feces. However, inorganic arsenic trioxide was excreted 
much more rapidly than fish-arsenic; also exclusively in the urine 
(Peoples, 1983). The difference in metabolism of arsenic between 
the two animals has been explained by the fact that rats accumulate 
arsenic in their red blood cells accounting for as much as 90% of all 
the arsenic stored (Zingara and Bottino, 1983). 

In mice, nearly 100% of the ingested arsenite and arsenate are 
absorbed in the gastrointestinal tract during the initial stages of 
exposure (Zingaro and Bottino, 1983). The retention of arsenic in the 
body is approximatey equal for both valence states at low doses (0.4 mg 
kg" 1 ), but at higher doses (4.0 mg kg ) , the retention time of arsenite 
exceeds that of arsenate. The metabolism of arsenite and arsenate differ 
with arsenite ingestion leading to significantly higher concentrations of 
arsenic in the liver and bile. However, both inorganic compounds are 
eventually converted into dimethylarsinic acid. 

Cows and dogs fed either arsenite or arsenate excreted 50% of the 
arsenic in the inorganic form while the remainder was metabolized and 
excreted as methylarsenic compounds. It was concluded that these animals 
rapidly synthesize organoarsenics and the methylation reaction was not 
dependent on intestinal microflora (Lakso and Peoples, 1975). Humans also 
convert inorganic arsenic into methylated species forming dimethylarsinic 



40 



acid and methanearsonic acid. In one study, approximately 25% of ingested 
arsenate was excreted in the urine within 24 hr and within 5 days the 
total reached 58°/. (Tarn et al., 1979). Of the excreted arsenic, 707. had 
been converted into methylated forms. Dimethylarsinic acid was the 
predominant species in urine after the initial 24 hr period and continued 
to increase in proportion relative to other arsenic compounds. In 
another study, 200 ug of arsenate in well water was imbibed and arsenate 
levels in the urine increased significantly up to 10 hr of exposure at 
which time dimethylarsinic acid became the predominate form of arsenic 
(Crecelius, 1977). 

Consumption of wine containing an equivalent of 50 ug of arsenite 
and 13 ug of arsenate led to the excretion of inorganic arsenic in urine 
reaching a maximum after 5-10 hr (Crecelius, 1987). The concentrations 
of dimethylarsinic acid and methanearsonic acid steadily increased but 
did not reach a maximum until 40 hr after consumption. Dimethylarsinic 
acid accounted for 50% of the total arsenic ingested. 

Inorganic arsenic administered as arsenic trioxide (As + ) was given 
to an individual as an oral dose of 700 ug (Yamauchi and Yamamura, 1979). 
After 12 hr, 40% of the total arsenic ingested was excreted in the urine 

and 70% had been excreted within 72 hr. The percentage of each arsenic 

+3 +5 
species excreted during the 72 hr period was 21.7% As , 4.9% As ; 19.1% 

methanearsonic acid and 19.6% dimethylarsinic acid. As + was rapidly 

excreted within 12 hr of ingestion and methylated arsenic compounds 

reached a maximum at 12 hr, staying relatively constant for an additional 

36 hr after which levels begin to decrease. 



41 



Seafood contains high concentrations of organic arsenic either as 
complex arsenosugars and arsenolipids or as arsenbetaine. Humans fed 
shrimp, fish, or crab meat containing arsenic rapidly excrete organo- 
arsenic compounds in their urine. Over 757. of ingested fish-arsenic is 
excreted within 8-9 days and less than 1% can be accounted for in feces 
(Tarn et al., 1982). Arsenobetaine has been identified in urine of humans 
fed lobster (Cannon et al., 1981). Ingestion of organoarsenic compounds 
from marine animals are rapidly absorbed and circulated into the blood 
stream where they are passed to the kidneys and excreted in the urine 
largely unchanged. 

Ingested methanearsonic acid and dimethylarsinic acid are more 
rapidly excreted from the body than arsenite. Dimethylarsinic acid is 
recovered unchanged and no evidence suggests conversion to inorganic 
arsenic. The majority of methanearsonic acid is also excreted unchanged 
but approximately 137. is converted into dimethylarsinic acid (Zingaro and 
Bottino, 1983). In rats, 248 min was required for absorption of 507. of 
the ingested dimethylarsinic acid by the gastrointestinal tract while 
only 2.2 min was required for the same level of absorption through the 
lungs. Dimethylarsinic acid like inorganic arsenic has a high affinity 
for rat erythrocytes and the concentration in all tissues decrease 
rapidly except in the blood where the half-life of arsenic parallels the 
half-life of the erythrocytes (Stevens et al., 1977). 

Inhalation of arsenic trioxide, As-O,, with smelter dust results 
in the absorption of arsenic into the body and its biotransformation 
into organic derivatives. Smith et al. (1977) reported that human urine 
samples contained 50 to 707. of arsenic as dimethylarsinic acid, 207. as 



42 



methanearsonic acid and the remainder as inorganic arsenite and arsenate 
after exposure to arsenic trioxide. Dimethylarsinic acid levels in the 
urine respond to small changes in airborne concentrations of arsenic 
trioxide. Urine concentrations of dimethylarsinic acid have been 
proposed as an indicator for monitoring airborne exposure to arsenic 
trioxide. 

In animals, inorganic arsenic is apparently removed by two 
processes. The first is the rapid absorption then assimilation into 
the blood followed by removal in the kidneys and passage from the body 
in the urine. The second process is much slower and involves the detoxi- 
fication of the inorganic compounds by the conversion to methylated forms 
(Crecelius, 1977). The methylated arsenic compounds are less toxic to 
animals and, in addition, the methylated forms are rapidly excreted from 
the body (Zingaro and Bottino, 1983). 

TOXICITY OF ARSENIC 

The toxicity of arsenic depends on the valency state. Arsenate 
inhibits ATP synthesis by uncoupling oxidative phosphorylation leading 
to the breakdown of energy metabolism (Craig, 1986). Arsenate may also 
replace phosphate in substituted monosaccharides such as glucose-6- 
phosphate yielding glucose-6-arsenate. Under standard conditions, 
arsenite is more toxic than arsenate, to aquatic organisms. Arsenite 
reacts with thiol groups present on active sites of many enzymes and 
tissue proteins such as keratin in skin, nails and hair (Schroeder and 
Balassa, 1966; Knowles and Benson, 1983). It covalently links to sulfur 



43 



atoms inactivating enzymes. In mammalian systems, arsenite has a longer 
half life than other arsenic species. Common symptoms of toxicity 
include chronic intoxication with decreased motor coordination, nervous 
disorders, respiratory distress and damage to kidneys and respiratory 
tract. 

The toxicity of arsenic to aquatic plants and invertebrates is a 
function of pH and usually decreases with increasing pH reflecting a 
change in oxidation states. In addition_, phosphate loading introduced 
to a culture medium produces an antagonistic effect on the toxicity of 
arsenic to aquatic plants. This is thought to be a result of competition 
of phosphate with arsenate for uptake. 

The toxicity of arsenic has been suggested to be the result of 
arsenate reduction to arsenite. Arsenate is excreted through urine 
more readily than arsenite because of its poor affinity for thiol groups. 
Methylation of inorganic arsenic j_n vivo has been reported in animals 
and humans with organoarsenic compounds being much less toxic than the 
inorganic forms. The LD 5Q for dimethylarsinic (cacodylic) acid in rats 
range from 700 to 2600 mg kg compared to methanearsonic acid (700 to 
1800 mg kg ) , potassium arsenite (14 mg kg" ) and calcium arsenate (20 
mg kg" ) (Craig, 1986). The arsenic analogues of choline and betaine 
are considered nontoxic and can be fed in high amounts (percent level) 
to animals. They do not have the ability to bind to thiol groups and 
are resistant to conversion to the more toxic forms. Organoarsenic 
compounds which accumulate in seafoods (fish, crustaceans, aquatic 
plants) are rapidly excreted in unchanged forms by animals and humans. 



44 



The volatile arsine gases (LD 5Q in rats; 3 mg kg" ) appear to be 
highly toxic to mammals inducing lysis of red blood cells. Although 
arsenic hydride (AshL) is extremely toxic, it is very unstable, and not 
typically found in nature. Arsines normally undergo alkylation and 
arylation reactions before being released into the environment. Further 
tests are needed to determine the toxicity of dimethylarsine and 
trimethylarsine with inhalation studies to test animals. 

STABILITY OF ORGANOARSENIC COMPOUNDS 

Both methanearsonic acid and dimethylarsinic acid are somewhat 
persistent in the environment. These compounds can be taken up by 
plants and not metabolized to any significant extent (Hiltbold, 1975). 
In mammals, methanearsonic acid may be methylated to dimethylarsinic 
acid, but direct demethylation of the latter has not been demonstrated 
(Stevens et al., 1977). Consumption of seafood (lobster, shrimp, 
crabmeat, flounder) containing organoarsenic compounds is excreted 
directly through the human body unchanged in chemical form. Uptake and 
digestion of organoarsenic compounds by waterfowl and wildlife near and 
within the agricultural evaporation ponds in the Tulare Lake Basin would 
also not be expected to be much of a threat. 

Demethylation of organoarsenic compounds in both natural waters 
and soil has been demonstrated and is apparently widespread among 
many bacteria. Achromobacter sp., Flavobacterium sp., Nocardia sp., 
Pseudomonas sp., and Al ical igenes sp., isolated from soil and sediment 
demethylate methanearsonic acid at a rate of 3-5% per 48 h (Shariatpanahi 



45 



et al., 1981). Methanearsonic acid is transformed to arsenate and carbon 
dioxide. 

GLOBAL CYCLE OF ARSENIC 

The global cycle of arsenic is illustrated in Fig. 5. Many organ- 
isms including microorganisms, plants and invertebrates are involved in 
the distribution and cycling of this element. Arsenic can accumulate and 
be subject to various transformations including reduction, oxidation and 
methylation. The reduced form (arsenite) is considered more toxic than 
the oxidized species (arsenate) because it reacts with sulfhydryl groups 
of cysteine in proteins inactivating many enzymes. 

In aquatic systems, arsenic tends to accumulate as complex organo- 
arsenic compounds with only a few being identified (e.g., arsenobetaine, 
arsenochloine and dimethylarsenosoribosides). Methanearsonic acid and 
dimethylarsinic acid are present in seawater and freshwater but appear to 
be degradation products of these complex organoarsenic compounds. 

Arsenic is emitted into the atmosphere by high temperature processes 
such as coal-fired power generation plants, burning vegetation and 
volcanism. Inputs into the atmosphere include industrial and fossil fuel 

Q _ 1 Q _ 1 

emission (780 x 10 g arsenic yr" ), mining (28 x 10 g arsenic yr" ) and 

8 1 

continental and volcanic dust fluxes (28 x 10 g arsenic yr" ) (Mackenzie 

et al., 1979). 

Natural low temperature biomethylation also releases arsenic into 

the atmosphere. Microorganisms including bacteria, fungi and yeast 

form volatile methylated derivatives of arsenic under both aerobic and 



46 




\ Fossi 

\ Fuels 

\ 158.3 

v x 
Mining of \ 

Sulfide Ores x 
And \ . 

Cement N 
Manufacturing 
621 



Fig. 5. Global biogeochemical cycle of arsenic, 
fluxes are expressed in units of 10 8 and 
(Mackenzie et al., 1979). 



Reservoir masses and 
1q8 g/yr, respectively 



47 



anaerobic conditions. Bacteria only produce dimethylarsine while fungi 
synthesize trimethylarsine. Dimethylarsine is an oxidation product of 
trimethylarsine and both compounds are subject to demethylation by soil 

g 

bacteria. It is estimated that as much as 210 x 10 g of arsenic is lost 
to the atmosphere in the vapor state annually from the land surface 
(Mackenzie et al., 1979). The continental vapor flux is about eight 
times that of the continental dust flux indicating that the biogenic 
contribution may play a significant role in cycling of arsenic. It has 
not been established whether volatile arsenic can be relased by plants as 
appears to be the case for selenium. 



48 



REFERENCES 



Andreae, M. 0. 1979. Arsenic speciation in seawater and interstitial 
waters: The influence of biological-chemical interactions on the 
chemistry of a trace element. Limnol. Oceanogr. 24(3):440-452. 

Andreae, M. 0., and D. Klumpp. 1979. Biosynthesis and release of 
organoarsenic compounds by marine algae. Environ. Sci. Technol. 
13:738-741. 

Asher, C. J., and P. F. Reay. 1979. Arsenic uptake by barley seedlings. 
Aust. J. Plant Physiol. 6:459-466. 

Baker, M. D., W. E. Inniss, C. I. Mayfield, P.T.S. Wong, and Y. K. Chau. 
1983. Effect of pH on the methylation of mercury and arsenic by 
sediment microorganisms. Environ. Technol. Lett. 4:89-100. 

Baker, M. D., P.T.S. Wong, Y. K. Chau, C. I. Mayfield, and W. E. Inniss. 
1983. Methylation of arsenic by freshwater green algae. Can. J. 
Fish. Aquat. Sci. 40:1245-1257. 

Benson, A. A., R. V. Cooney, and J. M. Herrera-Lasso. 1981. Arsenic 
metabolism in algae and higher plants. J. Plant Nutr. 3:285-292. 

Benson, A. A., and P. Nissen. 1982. The arsenolipids of aquatic plants. 
Developments in Plant Biol. 8:121-124. 

Benson, A. A., and R. E. Summons. 1981. Arsenic accumulation in Great 
Barrier Reef invertebrates. Science 211:482-483. 

Bird, M. L. , F. Challenger, P. T. Charlton, and J. 0. Smith. 1948. 
Studies of biological methylation. II. The action of moulds on 
inorganic and organic compounds of arsenic. Biochem. J. 43:78-83. 

Bottino, N. R., E. R. Cox, K. J. Irgolic, S. Aeda, W. J. McShane, R. A. 
Stockton, and R. A. Zingaro. 1978a. Arsenic uptake and metabolism 
by the alga Testraselmis Chin. In Organometals and Organometal loids 
Occurrence and Fate in the Environment (ed. Brinckman, F. E., and 
J. M. Bel lama). Am. Chem. Soc. Symp. Ser. 82:116-129. 

Bottino, N. R., R. D. Newman, E. R. Cox, R. A. Stockton, M. Hoban, R. A. 
Zingaro, and K. J. Irgolic. 1978b. The effects of arsenate and 
arsenite on the growth and morphology of the marine unicellular 
algae Tetraselmis Chui (Chlorophyta) and Hymenomonas carterae 
(Chrysophyta) . J. Exp. Mar. Biol. Ecol. 33:153-168. 

Bowen, H.J.M. 1979. Environmental Chemistry of the Elements. Academic 
Press, New York. 



49 



Braman, R. S., and C. C. Foreback. 1973. Methylated forms of arsenic in 
the environment. Science 182:1247-1249. 

Button, D. K., S. S. Dunker, and M. L. Morse. 1973. Continuous culture 
of Rhodotorula rubra : Kinetics of phosphate-arsenate uptake, 
inhibition, and phosphate-limited growth. J. Bacteriol. 113:599-611. 

Cannon, J. R., J. S. Edmonds, K. A. Francesconi, C. L. Raston, J. B. 

Saunders, B. W. Skelton, and A. H. White. 1981. Isolation, crystal 
structure and synthesis of arsenobetaine, a constituent of the 
Western rock lobster, the dusky shark, and some samples of human 
urine. Aust. J. Chem. 34:787-798. 

Challenger, F. 1945. Biological methylation. Chem. Reviews 36:315-361. 

Challenger, F., C. Higginbottom, and L. Ellis. 1933. The formation of 
organo-metalloid compounds by microorganisms. Part I. Trimethyl- 
arsine and dimethylethylarsine. J. Chem. Soc. 95-101. 

Chau, Y. K., and P.T.S. Wong. 1978. Occurrence of biological methylation 
of elements in the environment. Am. Chem. Soc. Symp. Ser. 82:39-53. 

Chen, C.-M., T. K. Misra, S. Silver, and B. P. Rosen. 1986. Nucleotide 
sequence of the structural genes for an anion pump. J. Biol. Chem. 
261:15030-38. 

Cheng, C. N., and D. D. Focht. 1979. Production of arsine and methyl- 
arsines in soil and in culture. Appl. Environ. Microbiol. 
38:494-498. 

Chilvers, D. C, and P. J. Peterson. 1987. Global cycling of arsenic. 
In: Lead, Mercury, Cadmium and Arsenic in the Environment (eds. 
T. C. Hutchinson and K. M. Meema. John Wiley and Sons Ltd. pp. 
279-301. 

Cooney, R. V., and A. A. Benson. 1980. Arsenic metabolism in Homarus 
americanus . Chemosphere 9:335-341. 

Coulson, E. J. 1935. Metabolism in the rat of the naturally occurring 
arsenic of shrimp as compared with arsenic trioxide. J. Nutr. 
10:255-270. 

Cox, D. P., and M. Alexander. 1973a. Production of trimethylarsine gas 
from various arsenic compounds by three sewage fungi. Bull, of 
Environ. Contam. Toxicol. 9:84-88. 

Cox, D. P., and M. Alexander. 1973b. Effect of phosphate and other 
anions on trimethylarsine formation by Candida humicola . Appl. 
Microbiol. 25:408-413. 



50 



Cox, D. P., and M. Alexander. 1974. Factors affecting trimethylarsine 
and dimethylselenide formation by Candida humicola . J. Microb. 
Ecol. 1:136-144. 

Craig, P. J. 1986. Organometal 1 ic compounds in the environment, pp. 
198-228. Longman, England. 

Crecelius, E. A. 1975. The geochemical cycle of arsenic in Lake 

Washington and its relation to other elements. Limnol. Oceanoogr. 
20:441-451. 

Crecelius, E. A. 1977. Changes in the chemical speciation of arsenic 

following ingestion by man. Environ. Health Perspectives 19:147-150. 

Crecelius, E. A., M. H. Bothner, and R. Carpenter. 1975. Geochemistries 
of arsenic, antimony, mercury and related elements in sediments of 
Puget Sound. Environ. Sci. Technol. 9:325-333. 

Cullen, W. R., B. C. McBride, and A. W. Pickett. 1979. The transforma- 
tion of arsenicals by Candida humicola Can. J. Microbiol. 25:1201- 
1205. 

Cullen, W. R., A. E. Erdman, B. C. McBride, and A. W. Pickett. 1983. 

The identification of dimethylphenylarsine as a microbial metabolite 
using a simple method of chemofocusing. J. Microbiol. Method. 
1:297-303. 

Cullen, W. R., C. L. Froese, A. Lui , B. C. McBride, D. J. Patmore, and 
M. Reimer. 1977. The aerobic methylation of arsenic by micro- 
organisms in the presence of L-methionine-methyl-d3. J. Organometal. 
Chem. 139:61-69. 

Cullen, W. R., B. C. McBride, A. W. Pickett, and J. Reglinski. 1984. The 
wood preservative chromated copper arsenate is a substrate for 
trimethylarsine biosynthesis. Appl. Environ. Microbiol. 47:443-444. 

Cullen, W. R., B. C. McBride, and J. Reglinski. 1984a. The reaction of 
methylarsenicals with thiols: Some biological implications. J. 
Inorg. Biochem. 21:179-194. 

Cullen, W. R., B. C. McBride, and J. Reglinski. 1984b. The reduction of 
trimethylarsine oxide to trimethylarsine by thiols: A mechanistic 
model for the biological reduction of arsenicals. J. Inorg. Biochem. 
21:45-60. 

Dabbs, E. A., and G. J. Sole. 1988. Plasmid-borne resistance to 
arsenate, cadmium and chloramphenicol in a Rhodococcus species. 
Mol. Gen. Genet. 211:148-154. 

Edmonds, J. S., and K. A. Francesconi. 1981. Isolation and identifica- 
tion of arsenobetaine from the American lobster Homarus americanus . 
Chemosphere 10:1041. 



51 



Edmonds, J. S., and K. A. Francesconi. 1981. Arseno-sugars from brown 
kelp ( Ecklonia radiata) as intermediates in cycling of arsenic in a 
marine ecosystem. Nature 289:602-604. 

Edmonds, J. S., and K. A. Francesconi. 1982. Isolation and crystal 

structure of an arsenic-containing sugar sulphate from the kidney of 
the giant clam, Tridacna maxima . X-ray crystal structure of (2S)-3- 
[5-Deoxy-5(dimethylarsinoyl )-B-D-ribofuranosyloxy]-2-hydroxyproply 
hydrogen sulphate. J. Chem. Soc. Perkin Trans. Vol. 1. pp. 2989-93. 

Edmonds, J. S., K. A. Francesconi, and J. A. Hansen. 1982. Dimethyl- 
oxarsylethanol from anaerobic decomposition of brown kelp ( Ecklonia 
radiata): A likely precursor of arsenobetaine in marine fauna. 
Experientia. 38:643. 

Edmonds, J. S., and K. A. Francesconi. 1983. Arsenic-containing 

ribofuranosides: Isolation from brown kelp Ecklonia radiata and 
nuclear magnetic resonance spectra. J. Chem. Soc. Perkin Trans. I. 
pp. 2375-2382. 

Edmonds, J. S., and K. A. Francesconi. 1987. Transformations of arsenic 
in the marine environment. Experientia. 43:553-557. 

Edmonds, J. S., K. A. Francesconi, J. R. Cannon, C. L. Raston, B. W. 

Skelton, and A. H. White. 1977. Isolation, crystal structure and 
synthesis of arsenobetaine, the arsenical constituent of the Western 
rock lobster Panul irus longipes cygnus George. Tetrahedron Lett. 
18:1543-1546. 

Ferguson, J. F., and J. Gavis. 1972. A review of the arsenic cycle in 
natural waters. Water Research 6:1259-1274. 

Fish, G. R. 1963. Observation on excessive weed growth in two lakes in 
New Zealand. N.Z. Jl. Bot. 1:410-418. 

Goldberg, S., and R. P. Glaubig. 1988. Anion sorption of a calcareous, 
montmorillonitic soil-arsenic. Soil Sci. Soc. Am. J. 52:1297-1300. 

Hanaoka, K., T. Matsumonto, S. Tagawa, and T. Kaise. 1987. Microbial 
degradation of arsenobetaine, the major water soluble organoarsenic 
compound occurring in marine animals. Chemosphere. 16:2545-2550. 

Hassler, R. A., D. A. Klein, and R. R. Meglen. 1984. Microbial contri- 
bution to soluble and volatile arsenic dynamics in retorted oil 
shale. J. Environ. Qual. 13:466-470. 

Hiltbold, A. E. 1979. Behavior of organoarsenicals in plants and soils. 
Am. Chem. Soc. Symp. Ser. 7:53-69. 



52 



Hood, R. D. and Associ. 1985. Cacodylic acid: Agricultural uses, 

biological effects, and environmental fate. Superintendent of 

Documents, U.S. Government Printing Office, Washington, D.C. pp. 
164. 

Howard, A. G., M. H. Arbab-Zavar, and S. Apte. 1982. Seasonal vari- 
ability of biological arsenic methylation in the estuary of the 
River Beaulieu. Marine Chem. 11:493-498. 

Irgolic, K. J., R. A. Stockton, and D. Charkaborti. 1983. Determination 
of arsenic compounds in water supplies, p. 282-305. In: W. H. 
Lederer and R. J. Fensterheim (ed.) Arsenic: Industrial, Biomedical, 
Environmental Perspectives, van Nostrand, New York. 

Johnson, D. L., and R. S. Braman. 1975. Alkyl and inorganic arsenic in 
air sarnies. Chemosphere. 4:333-338. 

Kaise, T., K. Hanaoka, and S. Tagawa. 1985. The formation of trimethyl- 
arsine oxide from arsenobetaine by biodegradation with marine 
microorganisms. Chemosphere. 16:2551-2558. 

Klumpp, D. W. 1980. Characteristics of arsenic accumulation by seaweed 
Fucus spiralis and Ascophyl lum nodosum . Marine Biol. 58:257-264. 

Klumpp, D. W., and P. J. Peterson. 1979. Arsenic and other trace 

elements in the waters and organisms of an estuary in S.W. England. 
Envir. Pollut. 19:11-20. 

Klumpp, D. W., and P. J. Peterson. 1981. Chemical characteristics of 
arsenic in a marine food chain. Marine Biol. 62:297-305. 

Knowles, F. C, and A. A. Benson. 1983. The biochemistry of arsenic. 
Trends in Biochem. Sci. 8:178-180. 

Kurosawa, S., K. Yasuda, M. Taguchi , S. Yamazaki, S. Toda, M. Morita, T. 
Uehiro, and K. Fuwa. 1980. Identification of arsenobetaine, a 
water soluble organo-arsenic compound in muscle and liver of a 
shark, Prionace glaucus . Agri. Biol. Chem. 44:1993-1994. 

Lakso, J. U., and S. A. Peoples. 1975. Methylation of inorganic arsenic 
by mammals. J. Agric. Food Chem. 23:674-676. 

LeBlanc, P. J., and A. L. Jackson. 1973. Arsenic in marine fish and 
invertebrates. Marine Pollut. Bull. 4:88-90. 

Lemmo, N. V., S. D. Faust, T. Belton, and R. Tucker. 1983. Assessment 
of the chemical and biological significance of arsenical compounds 
in a heavily contaminated watershed. Part I. The fate and 
speciation of arsenical compounds in aquatic environments - A 
literature review. J. Environ. Sci. Health, A18:335-387. 



53 



Letey, J., C. Roberts, M. Penberth, and C. Vasek. 1986. An agricultural 
dilemma: Drainage water and toxics disposal in San Joaquin Valley, 
Division of Agric. and Natural Resources, Univ. of California, 
Publication 3319. 

Lunde, G. 1972. The analysis of arsenic in the lipid phase from marine 
and limnetic algae. Acta Chem. Scandi. 26:2642-2644. 

Lunde, G. 1973. The synthesis of fat and water-soluble arseno-organic 
compounds in marine and limnetic algae. Acta Chem. Scand. 
27:1586-1594. 

Luten, J. B., G. Riekwel-Booy, J.v.d. Greef, M. C. ten Noever de Brauw. 
1983. Identification of arsenobetaine in sole, lemon sole, flounder, 
dab, crab, and shrimp by field desorption and fast atom bombardment 
mass spectrometry. Chemosphere 12:131-41. 

Mackenzie, F. T., R. J. Lantzy, and V. Paterson. 1979. Global trace 

metal cycles and predictions. J. Int. Assoc. Math. Geol. 11:99-142. 

Maher, W. A. 1983. Inorganic arsenic in marine organisms. Mar. Pol 1 u. 
Bull. 14:308-310. 

McBride, B. C, H. Merilees, W. R. Cullen, and W. Pickett. 1978. 

Anaerobic and aerobic alkylation of arsenic. In: Organometals and 
Organometal loids Occurrence and Fate in the Environment (ed. 
Brickman, F. E., and J. M. Bel lama). Am. Chem. Soc. Symp. Ser. 
82:94-115. 

McBride, B. C, and R. S. Wolfe. 1971. Biosynthesis of dimethylarsine 
by methanobacterium. Biochem. 10:4312-4317. 

Mehra, H. C, and W. T. Frankenberger, Jr. 1988. Single-column ion 
chromatography: IV. Determination of arsenate in soils. Soil Sci. 
Soc. Am. J. 52:1603-1606. 

Merril, W., and D. W. French. 1964. The production of arsenious gases 
by wood rotting fungi. Proc. Minn. Acad. Sci. 31:105-106. 

Mobley, H.L.T., C. Chen, S. Silver, and B. P. Rosen. 1983. Cloning and 
expression of R-factor mediated arsenate resistance in Escherichia 
coli . Mol. Gen. Genet. 191:421-426. 

Mobley, H. T., and B. P. Rosen. 1982. Energetics of plasmid-mediated 
arsenate resistance in Escherichia coli. Proc. Natl. Acad. Sci. 
79:6119-6122. 

Morita, M., T. Vehiro, and K. Fuwa. 1981. Determination of arsenic 
compounds in biological samples by HPLC with ICP detection. Anal. 
Chem. 53:1806-1808. 



54 



Morrison, J. L. 1969. Distribution of arsenic from poultry litter in 
broiler chickens, soil and crops. J. Agr. Food Chem. 17:1288-1290. 

Nakahara, H., T. Ishikawa, Y. Sarai , I. Kondo, H. Kozukue, and S. Silver. 
1977. Linkage of mercury, cadmium, and arsenate and drug resistance 
in clinical isolates of Pseudomonas aeruginosa . Appl. Environ. 
Microbiol. 33:975-976. 

Nissen, P., and A. A. Benson. 1982. Arsenic metabolism in freshwater 
and terrestrial plants. Physiol. Plant 54:446-450. 

Norin, H., and A. Christakopoulos. 1982. Evidence for the presence of 
arsenobetaine and another organoarsenical in shrimps. Chemosphere. 
11:287-298. 

Norin, H. R. Ryhage, A. Christakopoulos, and M. Sandstrom. 1983. New 
evidence for the presence of arsenocholine in shrimps ( Pandalus 
boreal is ) by use of pyrolysis gas chromatography — atomic absorption 
spectrometry/mass spectrometry. Chemosphere. 12:299-315. 

Onishi, H., and E. B. Sandell. 1955. Geochemistry of arsenic. Geochim. 
Cosmochim. Acta 7:1-33. 

Osborne, F. H., and H. L. Ehrlich. 1976. Oxidation of arsenite by a soil 
isolate of alcaligenes. J. Appl. Bacteriol. 41:295-305. 

Owens, J. W., and E. S. Gladney. 1976. The determination of arsenic in 
natural waters by flameless atomic absorption spectrometry. At. 
Absorpt. Newsl. 15:47-48. 

Penrose, W. R. 1975. Biosynthesis of organic arsenic compounds in brown 
trout (Sajmo trutta) . J. Fish. Res. Board Can. 32:2387-2390. 

Penrose, W. R., H.B.S. Conacher, R. Black, J. C. Meranger, W. Miles, H. M. 
Cunningham, and W. R. Squires. 1977. Implications of inorganic/ 
organic interconversion on fluxes of arsenic in marine food webs. 
Environ. Health Perspectives 19:53-59. 

Peoples, S. A. 1983. The metabolism of arsenic in man and animals. In: 
Arsenic: Industrial, Biomedical, Environmental Perspectives (ed. 
Lederer, W. H., and R. J. Fensterheim). Van Nostrand Reinhold, New 
York. pp. 125-133. 

Pepper, I., N. Galanti, J. Sans, and J. F. Lopez-Saez. 1988. Reversible 
inhibition of root growth and cell proliferation by pentavalent 
arsenic in AH i urn cepa L. Environ. Exper. Botany 28:9-18. 

Phillips, S. E., and M. L. Taylor. 1976. Oxidation of arsenite to 
arsenate by Alcaligenes faecalis . Appl. Environ. Microbiol. 
32:392-399. 



55 



Pickett, A. W., B. C. McBride, W. R. Cullen, and H. Manji. 1981. The 
reduction of trimethylarsine oxide by Candida humicola . Can. J. 
Microbiol. 27:773-778. 

Planas, D., and F. P. Healey. 1978. Effects of arsenate on growth and 
phosphorus metabolism of phytoplankton. J. Phycol. 14:337-341. 

Reay, P. F. 1972. The accumulation of arsenic from arsenic-rich natural 
waters by aquatic plants. J. Appl. Ecol. 9:557-565. 

Rosen, B. P., U. Weigel, C. Karkaria, and P. Gangola. 1988. Molecular 
characterization of an anion pump. J. Biol. Chem. 263:3067-3070. 

Rosenberg, H. R., R. G. Gerdes, and K. Chegwidden. 1977. Two systems 
for the uptake of phosphate in Escherichia col i . J. Bacterid . 
131:505-511. 

Sanders, J. G. 1979. The concentration and speciation of arsenic in 
marine macro-algae. Estuarine and Coastal Marine Sci. 9:95-99. 

Sanders, J. G. 1979. Effects of arsenic speciation and phosphate 
concentration of arsenic inhibition of Skeletonema costatum . J. 
Phycol. 15:424-428. 

Sanders, J. G. 1980. Arsenic cycling in marine systems. Marine Environ. 
Res. 3:257-266. 

Sanders, J. G., and H. L. Windom. 1980. The uptake and reduction of 
arsenic species by marine algae. Estuarine and Coastal Marine Sci. 
10:555-567. 

Schraufnagel , R. A. 1983. Arsenic in energy sources: A future supply or 
an environmental problem? In: Arsenic: Industrial, Biomedical, 
Environmental Perspectives (ed. Lederer, W. H., and R. J. 
Fensterheim). Van Nostrand Reinhold, New York. pp. 17-41. 

Schroeder, H. A., and J. J. Balassa. 1966. Abnormal trace metals in man: 
Arsenic. J. Chron. Dis. 19:85-106. 

Shariatpanahi , M., A. C. Anderson, and A. A. Abdelghani. 1981. Microbial 
demethylation of monosodium methanearsonate. In: Trace Substances 
in Environmental Health-X (ed. Hemphill, D. D.). University of 
Missouri, Columbia, pp. 383-387. 

Shiomi, K., A. Shinagawa, T. Igarashi, H. Yamanaka, and T. Kikuchi. 1984. 
Evidence for the presence of arsenobetaine as a major arsenic com- 
pound in the Shrimp Sergestes lucens . Experientia. 40:1247-1248.